CN111680421A - Home decoration design method and device, electronic equipment and storage medium - Google Patents
Home decoration design method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN111680421A CN111680421A CN202010506847.2A CN202010506847A CN111680421A CN 111680421 A CN111680421 A CN 111680421A CN 202010506847 A CN202010506847 A CN 202010506847A CN 111680421 A CN111680421 A CN 111680421A
- Authority
- CN
- China
- Prior art keywords
- designed
- sequence
- module
- information
- home decoration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013461 design Methods 0.000 title claims abstract description 99
- 238000005034 decoration Methods 0.000 title claims abstract description 94
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000003860 storage Methods 0.000 title claims abstract description 24
- 238000001514 detection method Methods 0.000 claims abstract description 31
- 238000010586 diagram Methods 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims description 13
- 230000007246 mechanism Effects 0.000 claims description 11
- 241001122767 Theaceae Species 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 9
- 238000007667 floating Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 238000000605 extraction Methods 0.000 description 6
- 230000003287 optical effect Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 238000005070 sampling Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 239000013307 optical fiber Substances 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000000717 retained effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 241000288105 Grus Species 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
Abstract
The embodiment of the invention discloses a home decoration design method, a home decoration design device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring a custom-type diagram to be designed; inputting the house type graph to be designed into a pre-trained detection network model to obtain room information of the house type to be designed; inputting the room information into a home decoration design model trained in advance to obtain home decoration design information of the house type to be designed; the home decoration design information comprises category information of at least one piece of furniture and placing position information of the furniture. The intelligent design of the home decoration is realized, the aim of automatically arranging the furniture is further realized, and the precision and the intelligent degree of the home decoration design are improved.
Description
Technical Field
The embodiment of the invention relates to the technical field of computers, in particular to a home decoration design method and device, electronic equipment and a storage medium.
Background
With the rapid development of the real estate industry, the market demand of home decoration design is rapidly increased, and the traditional home decoration design industry is a professional and time-consuming industry and cannot meet the demand of the market for rapid production. To meet the market demand, more and more enterprises increase the research, development and input of intelligent home decoration design.
At present, in the field of intelligent home decoration design, a design drawing is mainly obtained by matching based on a limited template of a part of rooms; automatic layout for furniture relies more on rule-based optimization techniques, which, however, are computationally inefficient and do not necessarily enable finding an optimal solution.
Therefore, the current intelligent home decoration design method needs to be further improved.
Disclosure of Invention
The embodiment of the invention provides a home decoration design method and device, electronic equipment and a storage medium, which realize intelligent design of home decoration, further realize the aim of automatically arranging furniture and improve the design precision and the intelligent degree of home decoration.
In a first aspect, an embodiment of the present invention provides a home decoration design method, where the method includes:
acquiring a custom-type diagram to be designed;
inputting the house type graph to be designed into a pre-trained detection network model to obtain room information of the house type to be designed;
inputting the room information into a home decoration design model trained in advance to obtain home decoration design information of the house type to be designed;
the home decoration design information comprises category information of at least one piece of furniture and placing position information of the furniture.
By designing a special detection network model for identifying the room information of the house type to be designed, the identification precision and the identification efficiency of the room information are improved, the workload of artificial marking is reduced, and the intelligent degree and the design speed of home decoration design are further improved. The home decoration design information comprises the category information of at least one type of furniture and the placing position information of the furniture, so that the furniture is automatically arranged without manually arranging the designed furniture.
Further, the detecting the network model comprises: EFGRNET (Enriched Feature guided refinement Network);
the EFGRNET comprises an SSD (Single Shot multi-box Detector) algorithm module, an FE (Feature Enriched) module and a cascade refinement module;
the FE Module includes an MSCF (Multi-Scale Contextual Feature) Module, and the cascade refinement Module includes an OM (Object Module) and an FGRM (Feature-guided refinement Module).
Further, the inputting the house type diagram to be designed into a pre-trained detection network model to obtain the room information of the house type to be designed includes:
processing the custom-designed graph to be designed into a first picture with a first set size and a second picture with a second set size, wherein the second set size is smaller than the first set size;
inputting the first picture into the SSD algorithm module so as to extract image features in the first picture through the SSD algorithm module;
inputting the second picture to the MSCF module to capture multi-scale contextual features in the second picture through the MSCF module;
and inputting the image features and the multi-scale context features into the cascade refinement module to obtain room information of the house type to be designed.
The backbone network of the SSD algorithm module is VGG (Visual Geometry Group), the image features in the first picture are extracted through the VGG network, and the network structure includes a plurality of convolution prediction layers for identifying context semantic information based on the image features, but feature information helpful for distinguishing a target region or a background region is still lost, for which, an FE module is introduced for capturing multi-scale context information. By designing the SSD algorithm module, the FE module and the cascade refinement module, the method not only improves the feature extraction speed, but also greatly improves the feature extraction precision, can obtain room information of a house type to be designed with higher precision, and provides a high-quality data source basis for intelligent home decoration design.
Further, the home decoration design model comprises: a first sequence-to-sequence Seq2Seq network and a second sequence-to-sequence Seq2Seq network;
the input of the first sequence to the sequence Seq2Seq network is a category sequence in room information of a house type to be designed, and the output is a category sequence of furniture;
the second sequence is input into the sequence Seq2Seq network as a coordinate sequence in the room information of the house type to be designed, and the second sequence is output as a category sequence of furniture;
by designing the first sequence to sequence Seq2Seq network and the second sequence to sequence Seq2Seq network, the simultaneous output of the furniture category and the furniture placing position is realized, the aim of automatically arranging the furniture is fulfilled, and the intelligent degree of home decoration design is improved.
Further, the first sequence-to-sequence Seq2Seq network and the second sequence-to-sequence Seq network each incorporate an attention mechanism for weighting a particular class of furniture based on functional characteristics of the room to ensure that furniture is designed for the room that matches the functional characteristics of the room. The reasonability and the usability of the home decoration design result are improved, namely, the home decoration design effect is improved, and the expectation and the use requirement of a user are met.
Further, before inputting the pattern of the house type to be designed into the pre-trained detection network, the method further includes:
and preprocessing the house pattern to be designed to obtain the data of the house pattern to be designed which meets the input requirement of the detection network.
The preprocessing comprises at least one of cropping, upsampling and downsampling;
the category information of the furniture includes at least one of: beds, desks, wardrobes, dressing tables, tea tables, and television cabinets;
the room information includes at least one of: the system comprises a door, a door position information, a wall body position information, a floating window position information and a window position information.
In a second aspect, an embodiment of the present invention further provides a home decoration designing apparatus, where the apparatus includes:
the device comprises a to-be-designed layout acquisition module, a to-be-designed layout acquisition module and a layout acquisition module, wherein the to-be-designed layout acquisition module is used for acquiring a to-be-designed layout;
the room information acquisition module is used for inputting the house type graph to be designed into a pre-trained detection network model to acquire the room information of the house type to be designed;
the design module is used for inputting the room information into a home decoration design model trained in advance to obtain home decoration design information of the house type to be designed;
the home decoration design information comprises category information of at least one piece of furniture and placing position information of the furniture.
In a third aspect, an embodiment of the present invention further provides an apparatus, where the apparatus includes:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement a method of designing a home decoration according to any one of the embodiments of the present invention.
In a fourth aspect, embodiments of the present invention further provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform a method for designing a home decoration according to any one of the embodiments of the present invention.
According to the technical scheme of the embodiment of the invention, the user-type diagram to be designed is obtained; inputting the house type graph to be designed into a pre-trained detection network model to obtain room information of the house type to be designed; inputting the room information into a home decoration design model trained in advance to obtain home decoration design information of the house type to be designed; the home decoration design information comprises category information of at least one piece of furniture and the technical means of the placement position information of the furniture, so that the intelligent design of home decoration is realized, the aim of automatic arrangement of the furniture is further realized, and the design precision and the intelligent degree of the home decoration are improved.
Drawings
The above and other features, advantages and aspects of various embodiments of the present invention will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and features are not necessarily drawn to scale.
Fig. 1 is a schematic flow chart of a home decoration design method according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a detection network model according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of a method for generating a home decoration design model according to a second embodiment of the present invention;
fig. 4 is a schematic structural diagram of a home decoration designing apparatus according to a third embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device according to a fourth embodiment of the present invention.
Detailed Description
Embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for a more thorough and complete understanding of the present invention. It should be understood that the drawings and the embodiments of the present invention are illustrative only and are not intended to limit the scope of the present invention.
It should be understood that the various steps recited in the method embodiments of the present invention may be performed in a different order and/or performed in parallel. Moreover, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the invention is not limited in this respect.
The term "include" and variations thereof as used herein are open-ended, i.e., "including but not limited to". The term "based on" is "based, at least in part, on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions for other terms will be given in the following description.
It should be noted that the terms "first", "second", and the like in the present invention are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.
It is noted that references to "a", "an", and "the" modifications in the present invention are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that reference to "one or more" unless the context clearly dictates otherwise.
Example one
Fig. 1 is a schematic flow chart of a home decoration design method according to an embodiment of the present invention, which is applicable to intelligent home decoration design of a house type to be designed, and eliminates the dependence on the professional of a home decoration designer, so as to achieve the purpose of generating a home decoration design drawing at a second level, and greatly improve the speed and the intelligent degree of home decoration design. The method may be performed by a home appliance design apparatus, which may be implemented in software and/or hardware.
As shown in fig. 1, the home decoration design method provided in this embodiment includes the following steps:
and step 110, acquiring a custom figure to be designed.
The house type graph to be designed can be a house type graph of a single room or a house type graph of all rooms of a set of houses. For example, the house type diagram may be a kitchen in a set of house, or may be a house type diagram of a single room, two rooms or three rooms including a kitchen, a living room, a bedroom, a bathroom and the like. The floor plan may be the most primitive floor plan, such as a blank floor plan.
The layout to be designed may be provided by the user in the form of an electronic picture.
And 120, inputting the house type graph to be designed into a pre-trained detection network model to obtain the room information of the house type to be designed.
Referring to fig. 2, a schematic structural diagram of a detection network model is shown, where the detection network model includes: the method comprises the steps of enriching feature-guided refinement network EFGRNET;
the EFGRNET comprises a single-pulse multi-box detector SSD algorithm module 210, a feature-rich FE module 220 and a cascade refinement module 230; the FE module 220 comprises a multi-scale contextual feature MSCF module and the cascade refinement module 230 comprises an objectification module and a feature-guided refinement module FGRM. The backbone network of the SSD algorithm module 210 is a VGG, and the image features in the first picture are extracted through the VGG network, and the network structure includes a plurality of convolution prediction layers for identifying context semantic information based on the image features, but feature information helpful for distinguishing a target region or a background region is still lost. By designing the SSD algorithm module, the FE module and the cascade refinement module, the method not only improves the feature extraction speed, but also greatly improves the feature extraction precision, can obtain room information of a house type to be designed with higher precision, and provides a high-quality data source basis for intelligent home decoration design.
Illustratively, the inputting the house type diagram to be designed into a pre-trained detection network model to obtain room information of the house type to be designed includes:
processing the custom-designed graph to be designed into a first picture with a first set size and a second picture with a second set size, wherein the second set size is smaller than the first set size;
inputting the first picture into the SSD algorithm module so as to extract image features in the first picture through the SSD algorithm module;
inputting the second picture to the MSCF module to capture multi-scale contextual features in the second picture through the MSCF module;
and inputting the image features and the multi-scale context features into the cascade refinement module to obtain room information of the house type to be designed.
For example, the first set size may be 320 × 320, and the second set size may be 40 × 40. Specifically, referring to fig. 2, the 320 × 320 first picture is input to the SSD algorithm module 210, the image features are extracted through the VGG network, then prediction is performed using the conv4_3, fc7, conv8_2, and conv9_2 prediction layers, and the predicted result is input to the cascade refinement module 230. Through four prediction layers, some degree of context semantic information may be retained, but feature information that helps to distinguish between a target region or a background region may still be lost. In this regard, the feature-rich FE module 220 is introduced into the detection network model provided in the present embodiment, and is used to capture multi-scale context information of the input picture. Specifically, the input user type graph to be designed is subjected to pooling downsampling operation to obtain a 40 × 40 second picture, and then the second picture is input into the MSCF module, wherein the MSCF module is mainly used for capturing multi-scale context features. The output of the MSCF module is input to the objectification module of the cascade tessellation module 230 along with the output of the prediction layer of the SSD algorithm module 210, and the output of the objectification module is input to the feature-guided tessellation module FGRM. The rich features captured above (i.e., the output of the MSCF module and the output of the prediction layer of the SSD algorithm module 210) are subjected to binary classification (C1x) and initial box regression (B1x) at each prediction layer, and then FGRM is used to solve the problem of imbalance between the positive and negative anchors, finally outputting the detected room information. By the EFGRNET detection network model, the purpose of extracting the multi-scale context features of the house type picture to be designed is achieved, the identification precision and the identification efficiency of room information are improved, the workload of artificial marking is reduced, and the intelligent degree and the design speed of home decoration design are further improved. The room information specifically includes: category information and coordinate information, such as at least one of: the system comprises a door, a door position information, a wall body position information, a floating window position information and a window position information. Based on the identified room information, the available space in the room of the house type to be designed for furniture placement can be determined.
Wherein the category information of the furniture includes at least one of: beds, desks, wardrobes, dressing tables, tea tables, and television cabinets.
Further, the home decoration design model comprises: a first sequence-to-sequence Seq2Seq network and a second sequence-to-sequence Seq2Seq network;
the input of the first sequence to the sequence Seq2Seq network is a category sequence in room information of a house type to be designed, and the output is a category sequence of furniture;
and the second sequence is input into the sequence Seq2Seq network as a coordinate sequence in the room information of the house type to be designed, and the second sequence is output as a category sequence of furniture.
The first sequence-to-sequence Seq2Seq network and the second sequence-to-sequence Seq network each incorporate an attention mechanism for weighting a particular class of furniture based on functional characteristics of the room to ensure that furniture is designed for the room that matches the functional characteristics of the room. For example, a room may be characterized as a restaurant, giving dining tables and chairs higher weight relative to other furniture; if the functional characteristic of the room is the living room, the sofa, the tea table, the television cabinet and the television are endowed with higher weight relative to other furniture, so that the dining table and the dining chair are at least designed for the dining room, and the sofa, the tea table, the television cabinet and the television are at least designed for the living room. Through the operation, the effect and the precision of the home decoration design can be further improved, so that the design result can be accepted by the user. The attention mechanism is a problem solving method which is proposed by simulating human attention, namely simply screening high-value information from a large amount of information quickly, and is mainly used for solving the problem that a final reasonable vector representation is difficult to obtain when a model input sequence is long. The Attention model is used in the decoding process, it changes the disadvantage that the traditional decoder assigns the same weight to each input, but assigns different weights according to the different input objects, in the encoding encoder process, the output is no longer a fixed length intermediate result, but a sequence composed of different length vectors, and the decoder process proceeds further processing according to the sequence subset.
It can be understood that, before inputting the pattern of the house to be designed into the pre-trained detection network, the method further includes:
and preprocessing the house pattern to be designed to obtain the data of the house pattern to be designed which meets the input requirement of the detection network. The pre-processing includes at least one of cropping, upsampling, and downsampling.
According to the technical scheme of the embodiment of the invention, the EFGRNET detection network model is used for detecting the room information to be designed, the aim is to detect hardware entities such as walls, doors, windows and bay windows in the room, so that the effective space for placing furniture in the room is determined based on the hardware entities in the room, the EFGRNET detection network model can be used for comprehensively identifying the hardware entities in the room, and the identification precision and efficiency are improved. By designing a home decoration design model comprising two Seq2Seq networks, one is used for determining the category sequence of furniture, and the other is used for determining the coordinate sequence of the furniture, the automatic arrangement of the furniture is realized, manual arrangement is not needed, the intelligent degree of home decoration design is improved, further, an attention mechanism is introduced into each Seq2Seq network, the attention mechanism is used for giving higher weight to necessary furniture, the purpose of improving the home decoration design to accord with the use habit of a user is achieved, and the home decoration design result is easier to be favored by the user.
Example two
Fig. 3 is a flowchart illustrating a method for generating a home decoration design model according to a second embodiment of the present invention. On the basis of the above embodiments, the present embodiment describes in detail the training process of detecting the network model and the home decoration design model. Wherein explanations of the same or corresponding terms as those of the above embodiments are omitted
As shown in fig. 3, the method comprises the steps of:
and 310, collecting and finishing decoration design drawings to obtain a training data set.
Specifically, the home decoration design drawings are collected and arranged, and the home decoration design drawings are mainly carried out by a home decoration designer aiming at specific house types. And marking the room information and the furniture arrangement information of the house type by using a marking tool. The room information is specifically position information of a door and the door, position information of a wall body and the wall body, position information of a floating window and the floating window, and position information of the window and the window. The labeled room information is used as a training data set for training the detection network model. The arrangement information of the furniture includes a type and a coordinate position of the furniture, and the type of the furniture includes, for example: beds, desks, chairs, television cabinets, wardrobes, and the like. And the marked furniture arrangement information is used as a training data set of the home decoration design model.
And step 320, constructing and training an EFGRNET network model.
Specifically, in order to improve the generalization ability of the model, the sample training set can be enriched by performing operations such as turning, clipping, and adjusting the brightness of the picture on the input picture. Before the house type picture is input into the EFGRNET network model, the house type picture is subjected to fuzzy processing, and the aim is to prevent the house type picture from being not clear enough and influencing the prediction effect of the model. Meanwhile, the size of the processed house type pictures is processed into 320 × 320, the size of the pictures is processed into 40 × 40 through down sampling, and the pictures with the sizes of 320 × 320 and 40 × 40 are input into the EFGRNET network model. As shown in fig. 2, the EFGRNET includes a single-pulse multi-box detector SSD algorithm module 210, a feature-rich FE module 220, and a cascade refinement module 230; the FE module 220 comprises a multi-scale contextual feature MSCF module and the cascade refinement module 230 comprises an objectification module and a feature-guided refinement module FGRM. The SSD algorithm module 210 uses the VGG network as the backbone network, inputs 320 × 320 picture data into the model, extracts image features through the VGG network, and performs prediction using the conv4_3, fc7, conv8_2, and conv9_2 prediction layers. The predicted result is input to the cascade refinement module 230. Through four prediction layers, some degree of context semantic information may be retained, but feature information that helps to distinguish between a target region or a background region may still be lost. To address this issue, the feature-rich FE module 220 is used to capture multi-scale contextual information of the input picture. Specifically, pictures of size 40 × 40 are input into the MSCF module, which is mainly used to capture multi-scale contextual features. The output of the MSCF module is input to the objectification module of the cascade tessellation module 230 along with the output of the prediction layer of the SSD algorithm module 210, and the output of the objectification module is input to the feature-guided tessellation module FGRM. The rich features captured above (i.e., the output of the MSCF module and the output of the prediction layer of the SSD algorithm module 210) are subjected to binary classification (C1x) and initial box regression (B1x) at each prediction layer, and FGRM is then used to solve the imbalance problem between the positive and negative anchors, and finally output the detected room information, specifically the hardware label and coordinates (specifically the upper left and lower right coordinates of the target box).
And step 330, constructing and training a home decoration design model.
Dividing room information output by an EFGRNET network model into a hardware label sequence and a coordinate sequence, wherein the hardware label specifically comprises: the tag sequence is processed into sequence data E { E1, E2.. En } according to the clockwise direction, and simultaneously, identifiers of < s > and < E > respectively representing the beginning and the end of the sequence are added at the beginning and the end. Similarly, for the coordinate sequence, processing into the corresponding coordinate sequence according to the label order, it is also necessary to add [0,0,0,0] and [1,1,1,1] as the start and end identifiers. The same process is performed on the previously labeled class label and coordinate data of the furniture.
Two sequence2sequence network models are constructed, and GRUs can be used as basic models. The first sequence2sequence network model is used for training label sequence data, inputting hardware entity label training labels which are room information, and outputting class label sequences which are furniture. The second sequence2sequence network model is used to train coordinate sequence data, input as the coordinates of the hardware entities of the room, and output as the coordinate sequences of furniture. In both models, an attention mechanism is introduced, higher weights are respectively given to input doors and windows, different weights are given to different furniture in output furniture according to different functional areas, for example, a guest-dining room is given to a sofa, a tea table, a television cabinet, a television, a dining table and a dining chair, higher weights are given to the sofa, the tea table, the television cabinet, the television, the dining table and the dining chair relative to other furniture, and the functions of the sofa, the tea table, the television cabinet, the television, the dining table and the dining chair are ensured to be output when training output is carried out in the functional area of the guest-dining room, so that the design effect of.
It can be understood that the trained network model can be deployed to a server, and platforms with home decoration design requirements can share the network model at the server end, and specifically, a service can be provided to a user in an interface mode.
According to the technical scheme of the embodiment of the invention, the EFGRNET comprises a single-pulse multi-box detector SSD algorithm module, a characteristic-rich FE module and a cascade refinement module, so that the characteristic extraction speed is increased, the characteristic extraction precision is greatly increased, the room information of a house type to be designed with higher precision can be obtained, and a high-quality data source basis is provided for intelligent home decoration design; two sequence2sequence network models are designed in the home appliance model, one sequence2sequence network model is used for training label sequence data, inputting hardware entity label training labels which are room information and outputting category label sequences of furniture, the other sequence2sequence network model is used for training coordinate sequence data, inputting coordinates of hardware entities of rooms and outputting coordinate sequences of furniture, and the two sequence2sequence networks introduce an attention mechanism for giving weights to furniture of specific categories based on functional characteristics of the rooms so as to ensure that furniture matched with the functional characteristics of the rooms is designed for the rooms.
EXAMPLE III
Fig. 4 is a home decoration designing apparatus according to a third embodiment of the present invention, including: a to-be-designed house type map obtaining module 410, a room information obtaining module 420 and a design module 430.
The user type graph to be designed acquiring module 410 is configured to acquire a user type graph to be designed; a room information obtaining module 420, configured to input the house type diagram to be designed to a pre-trained detection network model, so as to obtain room information of the house type to be designed; the design module 430 is configured to input the room information into a pre-trained home decoration design model to obtain home decoration design information of the house type to be designed; the home decoration design information comprises category information of at least one piece of furniture and placing position information of the furniture.
On the basis of the above technical solutions, the detecting network model includes: the method comprises the steps of enriching feature-guided refinement network EFGRNET;
the EFGRNET comprises a single-pulse multi-box detector SSD algorithm module, a feature-rich FE module and a cascade refinement module;
the FE module comprises a multi-scale contextual feature MSCF module, and the cascade refinement module comprises an objectification module OM and a feature-guided refinement module FGRM.
On the basis of the above technical solutions, the room information acquiring module 420 includes:
the processing unit is used for processing the custom-designed graph to be designed into a first picture with a first set size and a second picture with a second set size, and the second set size is smaller than the first set size;
the first input unit is used for inputting the first picture to the SSD algorithm module so as to extract image features in the first picture through the SSD algorithm module;
a second input unit, configured to input the second picture to the MSCF module, so as to capture, by the MSCF module, multi-scale context features in the second picture;
and the third input unit is used for inputting the image characteristics and the multi-scale context characteristics to the cascade refinement module to obtain the room information of the house type to be designed.
On the basis of the above technical solutions, the home decoration design model includes: a first sequence-to-sequence Seq2Seq network and a second sequence-to-sequence Seq2Seq network;
the input of the first sequence to the sequence Seq2Seq network is a category sequence in room information of a house type to be designed, and the output is a category sequence of furniture;
and the second sequence is input into the sequence Seq2Seq network as a coordinate sequence in the room information of the house type to be designed, and the second sequence is output as a category sequence of furniture.
On the basis of the above technical solutions, attention mechanisms are introduced into the first sequence-to-sequence Seq2Seq network and the second sequence-to-sequence Seq2Seq network, and are used for giving weights to furniture of specific categories based on functional features of rooms, so as to ensure that furniture matched with the functional features of the rooms is designed for the rooms.
On the basis of the above technical solutions, the apparatus further includes:
and the preprocessing module is used for preprocessing the house type graph to be designed before inputting the house type graph to be designed into a pre-trained detection network so as to obtain the data of the house type graph to be designed according with the input requirement of the detection network.
On the basis of the technical schemes, the preprocessing comprises at least one of cutting, up-sampling and down-sampling;
the category information of the furniture includes at least one of: beds, desks, wardrobes, dressing tables, tea tables, and television cabinets;
the room information includes at least one of: the system comprises a door, a door position information, a wall body position information, a floating window position information and a window position information.
According to the technical scheme of the embodiment of the invention, the EFGRNET detection network model is used for detecting the room information to be designed, the aim is to detect hardware entities such as walls, doors, windows and bay windows in the room, so that the effective space for placing furniture in the room is determined based on the hardware entities in the room, the EFGRNET detection network model can be used for comprehensively identifying the hardware entities in the room, and the identification precision and efficiency are improved. By designing a home decoration design model comprising two Seq2Seq networks, one is used for determining the category sequence of furniture, and the other is used for determining the coordinate sequence of the furniture, the automatic arrangement of the furniture is realized, manual arrangement is not needed, the intelligent degree of home decoration design is improved, further, an attention mechanism is introduced into each Seq2Seq network, the attention mechanism is used for giving higher weight to necessary furniture, the purpose of improving the home decoration design to accord with the use habit of a user is achieved, and the home decoration design result is easier to be favored by the user.
The home decoration design device provided by the embodiment of the invention can execute the home decoration design method provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
It should be noted that, the units and modules included in the apparatus are merely divided according to functional logic, but are not limited to the above division as long as the corresponding functions can be implemented; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the embodiment of the invention.
Example four
Referring now to fig. 5, a schematic diagram of an electronic device (e.g., the terminal device or server of fig. 5) 400 suitable for implementing embodiments of the present invention is shown. The terminal device in the embodiments of the present invention may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet computer), a PMP (portable multimedia player), a vehicle terminal (e.g., a car navigation terminal), and the like, and a fixed terminal such as a digital TV, a desktop computer, and the like. The electronic device shown in fig. 5 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.
As shown in fig. 5, the electronic device 400 may include a processing means (e.g., a central processing unit, a graphics processor, etc.) 401 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)402 or a program loaded from a storage device 406 into a Random Access Memory (RAM) 403. In the RAM 403, various programs and data necessary for the operation of the electronic apparatus 400 are also stored. The processing device 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input/output (I/O) interface 405 is also connected to bus 404.
Generally, the following devices may be connected to the I/O interface 405: input devices 406 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 407 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage devices 406 including, for example, magnetic tape, hard disk, etc.; and a communication device 409. The communication means 409 may allow the electronic device 400 to communicate wirelessly or by wire with other devices to exchange data. While fig. 5 illustrates an electronic device 400 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided.
In particular, according to an embodiment of the present invention, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, an embodiment of the invention includes a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program containing program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication means 409, or from the storage means 406, or from the ROM 402. The computer program performs the above-described functions defined in the methods of embodiments of the invention when executed by the processing apparatus 401.
The terminal provided by the embodiment of the invention and the home decoration design method provided by the embodiment belong to the same inventive concept, technical details which are not described in detail in the embodiment of the invention can be referred to the embodiment, and the embodiment of the invention and the embodiment have the same beneficial effects.
EXAMPLE five
An embodiment of the present invention provides a computer storage medium, on which a computer program is stored, and the program, when executed by a processor, implements the home decoration design method provided by the above-described embodiment.
It should be noted that the computer readable medium of the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor device, apparatus, or a combination of any of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution apparatus, device, or apparatus. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution apparatus, device, or apparatus. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
In some embodiments, the clients, servers may communicate using any currently known or future developed network protocol, such as HTTP (HyperText transfer protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.
The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to:
acquiring a custom-type diagram to be designed;
inputting the house type graph to be designed into a pre-trained detection network model to obtain room information of the house type to be designed;
inputting the room information into a home decoration design model trained in advance to obtain home decoration design information of the house type to be designed;
the home decoration design information comprises category information of at least one piece of furniture and placing position information of the furniture.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including but not limited to an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present invention may be implemented by software or hardware. Where the name of a cell does not in some cases constitute a limitation on the cell itself, for example, an editable content display cell may also be described as an "editing cell".
The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), systems on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), and the like.
In the context of the present invention, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing description is only exemplary of the preferred embodiments of the invention and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other embodiments in which any combination of the features described above or their equivalents is encompassed without departing from the spirit of the disclosure. For example, the above features and (but not limited to) features having similar functions disclosed in the present invention are mutually replaced to form the technical solution.
Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the invention. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
Claims (10)
1. A home decoration design method, comprising:
acquiring a custom-type diagram to be designed;
inputting the house type graph to be designed into a pre-trained detection network model to obtain room information of the house type to be designed;
inputting the room information into a home decoration design model trained in advance to obtain home decoration design information of the house type to be designed; the home decoration design information comprises category information of at least one piece of furniture and placing position information of the furniture.
2. The method of claim 1, wherein the detecting the network model comprises: the method comprises the steps of enriching feature-guided refinement network EFGRNET;
the EFGRNET comprises a single-pulse multi-box detector SSD algorithm module, a feature-rich FE module and a cascade refinement module;
the FE module comprises a multi-scale contextual feature MSCF module, and the cascade refinement module comprises an objectification module OM and a feature-guided refinement module FGRM.
3. The method according to claim 2, wherein the inputting the pattern of the house type to be designed into a pre-trained detection network model to obtain the room information of the house type to be designed comprises:
processing the custom-designed graph to be designed into a first picture with a first set size and a second picture with a second set size, wherein the second set size is smaller than the first set size;
inputting the first picture into the SSD algorithm module so as to extract image features in the first picture through the SSD algorithm module;
inputting the second picture to the MSCF module to capture multi-scale contextual features in the second picture through the MSCF module;
and inputting the image features and the multi-scale context features into the cascade refinement module to obtain room information of the house type to be designed.
4. The method of claim 1, wherein the home appliance design model comprises: a first sequence-to-sequence Seq2Seq network and a second sequence-to-sequence Seq2Seq network;
the input of the first sequence to the sequence Seq2Seq network is a category sequence in room information of a house type to be designed, and the output is a category sequence of furniture;
and the second sequence is input into the sequence Seq2Seq network as a coordinate sequence in the room information of the house type to be designed, and the second sequence is output as a category sequence of furniture.
5. The method of claim 4, wherein the first sequence-to-sequence Seq2Seq network and the second sequence-to-sequence Seq network each incorporate an attention mechanism for weighting a particular class of furniture based on functional characteristics of the room to ensure that furniture is designed for the room that matches the functional characteristics of the room.
6. The method according to any one of claims 1-5, wherein before inputting the pattern of the house to be designed into the pre-trained detection network, the method further comprises:
and preprocessing the house pattern to be designed to obtain the data of the house pattern to be designed which meets the input requirement of the detection network.
7. The method of claim 6, wherein the pre-processing comprises at least one of cropping, upsampling, and downsampling;
the category information of the furniture includes at least one of: beds, desks, wardrobes, dressing tables, tea tables, and television cabinets;
the room information includes at least one of: the system comprises a door, a door position information, a wall body position information, a floating window position information and a window position information.
8. A home decoration designing apparatus, comprising:
the device comprises a to-be-designed layout acquisition module, a to-be-designed layout acquisition module and a layout acquisition module, wherein the to-be-designed layout acquisition module is used for acquiring a to-be-designed layout;
the room information acquisition module is used for inputting the house type graph to be designed into a pre-trained detection network model to acquire the room information of the house type to be designed;
the design module is used for inputting the room information into a home decoration design model trained in advance to obtain home decoration design information of the house type to be designed;
the home decoration design information comprises category information of at least one piece of furniture and placing position information of the furniture.
9. An electronic device, characterized in that the electronic device comprises:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the home design method of any one of claims 1-7.
10. A storage medium containing computer-executable instructions for performing the home appliance design method of any one of claims 1-7 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010506847.2A CN111680421A (en) | 2020-06-05 | 2020-06-05 | Home decoration design method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010506847.2A CN111680421A (en) | 2020-06-05 | 2020-06-05 | Home decoration design method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111680421A true CN111680421A (en) | 2020-09-18 |
Family
ID=72453955
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010506847.2A Pending CN111680421A (en) | 2020-06-05 | 2020-06-05 | Home decoration design method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111680421A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113449372A (en) * | 2021-07-20 | 2021-09-28 | 杭州群核信息技术有限公司 | Home decoration design method and device, computer readable storage medium and processor |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108984904A (en) * | 2018-07-17 | 2018-12-11 | 北京理工大学 | A kind of Home Fashion & Design Shanghai method based on deep neural network |
| CN109712248A (en) * | 2018-12-25 | 2019-05-03 | 北京惠佳家品科技有限公司 | A kind of 3D house ornamentation design platform system Internet-based |
| KR20190106867A (en) * | 2019-08-27 | 2019-09-18 | 엘지전자 주식회사 | An artificial intelligence apparatus for guiding arrangement location of furniture and operating method thereof |
| CN110348447A (en) * | 2019-06-27 | 2019-10-18 | 电子科技大学 | A kind of multiple-model integration object detection method with rich space information |
| CN110751721A (en) * | 2019-12-24 | 2020-02-04 | 广东博智林机器人有限公司 | Furniture layout drawing generation method and device, computer equipment and storage medium |
-
2020
- 2020-06-05 CN CN202010506847.2A patent/CN111680421A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108984904A (en) * | 2018-07-17 | 2018-12-11 | 北京理工大学 | A kind of Home Fashion & Design Shanghai method based on deep neural network |
| CN109712248A (en) * | 2018-12-25 | 2019-05-03 | 北京惠佳家品科技有限公司 | A kind of 3D house ornamentation design platform system Internet-based |
| CN110348447A (en) * | 2019-06-27 | 2019-10-18 | 电子科技大学 | A kind of multiple-model integration object detection method with rich space information |
| KR20190106867A (en) * | 2019-08-27 | 2019-09-18 | 엘지전자 주식회사 | An artificial intelligence apparatus for guiding arrangement location of furniture and operating method thereof |
| CN110751721A (en) * | 2019-12-24 | 2020-02-04 | 广东博智林机器人有限公司 | Furniture layout drawing generation method and device, computer equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 张婷 等: "《品质与创新理念下的产品设计研究》", 北京：中国书籍出版社, pages: 214 - 215 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113449372A (en) * | 2021-07-20 | 2021-09-28 | 杭州群核信息技术有限公司 | Home decoration design method and device, computer readable storage medium and processor |
| CN113449372B (en) * | 2021-07-20 | 2023-08-08 | 杭州群核信息技术有限公司 | Home decoration design method, apparatus, computer-readable storage medium, and processor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106651877B (en) | Instance partitioning method and device | |
| CN111898696A (en) | Method, device, medium and equipment for generating pseudo label and label prediction model | |
| CN111553012A (en) | Home decoration design method and device, electronic equipment and storage medium | |
| CN110189246B (en) | Image stylization generation method and device and electronic equipment | |
| CN111651819A (en) | Home decoration design method and device, electronic equipment and storage medium | |
| CN110209658B (en) | Data cleaning method and device | |
| CN111638791B (en) | Virtual character generation method and device, electronic equipment and storage medium | |
| CN109978077B (en) | Visual recognition method, device and system and storage medium | |
| CN111680123B (en) | Training method and device for dialogue model, computer equipment and storage medium | |
| CN114492831B (en) | Method and device for generating federal learning model | |
| CN111078940B (en) | Image processing method, device, computer storage medium and electronic equipment | |
| CN113408566A (en) | Target detection method and related equipment | |
| CN112861575A (en) | Pedestrian structuring method, device, equipment and storage medium | |
| CN111107278B (en) | Image processing method and device, electronic equipment and readable storage medium | |
| CN116934907A (en) | Image generation method, device and storage medium | |
| CN113033677A (en) | Video classification method and device, electronic equipment and storage medium | |
| CN110413820B (en) | Method and device for acquiring picture description information | |
| CN111652675A (en) | Display method and device and electronic equipment | |
| CN111612075A (en) | Interest point and descriptor extraction method based on joint feature recombination and feature mixing | |
| CN112446214B (en) | Advertisement keyword generation method, device, equipment and storage medium | |
| CN111680421A (en) | Home decoration design method and device, electronic equipment and storage medium | |
| CN110197459B (en) | Image stylization generation method and device and electronic equipment | |
| CN114757247A (en) | Training method of classification prediction model, classification prediction method, device and equipment | |
| CN112766406A (en) | Article image processing method and device, computer equipment and storage medium | |
| CN111128131A (en) | Voice recognition method and device, electronic equipment and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200918 |