WO2003088641A2 - 2:2 and 3:2 pull-down detection techniques - Google Patents

2:2 and 3:2 pull-down detection techniques Download PDF

Info

Publication number
WO2003088641A2
WO2003088641A2 PCT/US2003/010858 US0310858W WO03088641A2 WO 2003088641 A2 WO2003088641 A2 WO 2003088641A2 US 0310858 W US0310858 W US 0310858W WO 03088641 A2 WO03088641 A2 WO 03088641A2
Authority
WO
WIPO (PCT)
Prior art keywords
video signal
interlaced video
pull
generating
signal
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.)
Ceased
Application number
PCT/US2003/010858
Other languages
English (en)
French (fr)
Other versions
WO2003088641A3 (en
Inventor
Wing-Chi Chow
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zoran Corp
Original Assignee
Zoran Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zoran Corp filed Critical Zoran Corp
Priority to EP03746672A priority Critical patent/EP1510069A4/en
Priority to JP2003585418A priority patent/JP4533632B2/ja
Publication of WO2003088641A2 publication Critical patent/WO2003088641A2/en
Anticipated expiration legal-status Critical
Publication of WO2003088641A3 publication Critical patent/WO2003088641A3/en
Ceased legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/01—Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level
    • H04N7/0112—Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level one of the standards corresponding to a cinematograph film standard
    • H04N7/0115—Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level one of the standards corresponding to a cinematograph film standard with details on the detection of a particular field or frame pattern in the incoming video signal, e.g. 3:2 pull-down pattern
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/01—Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level
    • H04N7/0117—Conversion of standards, e.g. involving analogue television standards or digital television standards processed at pixel level involving conversion of the spatial resolution of the incoming video signal
    • H04N7/012—Conversion between an interlaced and a progressive signal

Definitions

  • the present invention relates to video signal processing and, more particularly, to improved 2:2 and 3:2 pull-down detection techniques.
  • the techniques of the present invention can, for example, be used when converting an interlaced video signal into a progressive video signal.
  • Interlaced display systems use interlaced video signals.
  • An interlaced video signal includes even fields, which contain one half of the total lines displayed on a display, and odd fields, which contain the other half of the total lines displayed on the display. The even fields and the odd fields of the interlaced video signal are alternately scanned onto the display to generate an image.
  • progressive display systems use progressive video signals.
  • a progressive video signal includes frames, each of which contains all of the lines displayed on a display. The frames of the progressive video signal are successively scanned onto the display to generate an image.
  • Video deinterlacing is also referred to as line doubling and video deinterlacer circuits are also referred to as line doubler circuits.
  • Two common video deinterlacing techniques are referred to as the merging technique (also referred to as weaving) and the interpolation technique (also referred to as bobbing).
  • the merging technique the lines of the even and odd fields of an interlaced video signal are weaved (or interleaved) to generate a single frame.
  • the merging technique is well suited for relatively static images, but produces highly objectionable artifacts when significant motion is present in the image.
  • the interpolation technique the interpolated lines (i.e., the missing lines) between the field lines are generated (usually by averaging the field pixels in the field lines above and below each interpolated line) and combined with the field lines to generate a single frame.
  • the interpolation technique is well suited for video with high motion content, but produces a clearly visible loss of vertical resolution for relatively static images.
  • Motion adaptive techniques have been developed so that when there is relatively little motion in an image, the merging technique is used, and when there is a relatively large amount of motion in an image, the interpolation technique is used.
  • Interlaced video signals can have different field rates, such as 50 fields/second or 60 fields/second, and interlaced video signals can be generated from different sources, such as film, cartoons, computer graphics, or computer animation.
  • Film which includes 24 frames that are displayed every second, is typically converted into a 50 field/second interlaced video signal using a well-known technique referred to as 2:2 pull-down and is then displayed at a rate 4% faster than the original rate.
  • Film is typically converted into a 60 field/second interlaced video signal using a well-known technique referred to as 3:2 pull-down (also referred to as 2:3 pull-down).
  • Computer animation is often created at 30 frames/second and is converted into 60 interlaced fields/second using 2:2 pull-down.
  • FIG. 1 illustrates the 2:2 pull-down technique.
  • film frame A is converted into even interlaced video field Al and odd interlaced video field A2;
  • film frame B is converted into even interlaced video field Bl and odd interlaced video field B2;
  • film frame C is converted into even interlaced video field CI and odd interlaced video field C2;
  • film frame D is converted into even interlaced video field Dl and odd interlaced video field D2, and so on.
  • 24 film frames are converted into 50 interlaced video fields every second.
  • Conventional video deinterlacer systems use a 2:2 pull-down field motion detector circuit to determine whether an interlaced video signal was generated using 2:2 pull-down.
  • the 2:2 pull-down field motion detector circuit compares consecutive fields of the interlaced video signal and generates a comparison value for each comparison. If the 2:2 pull-down field motion detector circuit detects a 2:2 pull-down field pattern, then the 2:2 pull-down field motion detector circuit generates a signal that indicates that the interlaced video signal was generated using 2:2 pull-down.
  • a 2:2 pull-down field pattern is a repeating sequence of 1 small comparison value followed by 1 large comparison value. Another signal is generated that indicates the 2:2 pull-down sequence.
  • the 2:2 pull-down field motion detector circuit determines whether the 2:2 pull-down field motion detector circuit has a good correlation between consecutive fields and a large comparison value indicates a poor correlation between consecutive fields.
  • a comparison value is the difference between two fields, and a difference value is the difference between two comparison values.
  • Fields Al and A2 were generated from film frame A; fields Bl and B2 were generated from film frame B; fields CI and C2 were generated from film frame C; fields Dl and D2 were generated from film frame D, and so on.
  • the 2:2 pull-down field motion detector circuit compares consecutive fields of the interlaced video signal yielding a repeating pattern of 1 small comparison value followed by 1 large comparison value:
  • A2 - Bl CV2 (which is a large difference)
  • B 1 - B2 CV3 (which is a small difference)
  • the 2:2 pull-down field motion detector circuit detects the repeating 2-field sequence of 1 small comparison value followed by 1 large comparison value and generates a signal which indicates that the interlaced video signal was generated using 2:2 pull-down. If a repeating 2-field sequence of 1 small comparison value followed by 1 large comparison value is not detected, the 2:2 pull-down field motion detector circuit generates a signal which indicates that the interlaced video signal was not generated using 2:2 pull-down.
  • the signal that indicates whether or not the interlaced video signal was generated using 2:2 pull-down is then provided to a deinterlacer circuit. If the indication signal indicates that a 2:2 pull-down field pattern has been detected, the deinterlacer circuit uses the merging technique to convert the interlaced video signal into a progressive video signal. On the other hand, if the indication signal indicates that a 2:2 pull-down field pattern has not been detected, the deinterlacer circuit typically uses a motion adaptive technique to convert the interlaced video signal into a progressive video signal.
  • deinterlacer circuit merges the consecutive fields in the interlaced video signal when a 2:2 pull-down field pattern is falsely detected. Since the consecutive fields in the interlaced video signal are from different points in time and are merged, highly objectionable artifacts (i.e., feathering or combing) appear in the displayed image.
  • FIG. 2 illustrates the 3:2 pull-down technique.
  • film frame A is converted into even interlaced video field Al, odd interlaced video field A2, and even interlaced video field A3 which is identical to field Al;
  • film frame B is converted into odd interlaced video field Bl and even interlaced video field B2;
  • film frame C is converted into odd interlaced video field CI, even interlaced video field C2, and odd interlaced video field C3 which is identical to field CI;
  • film frame D is converted into even interlaced video field Dl and odd interlaced video field D2, and so on.
  • 3:2 pull-down 24 film frames are converted into 60 interlaced video fields every second.
  • Conventional video deinterlacer systems use a 3:2 pull-down frame motion detector circuit to determine whether an interlaced video signal was generated using 3:2 pull-down.
  • the 3:2 pull-down frame motion detector circuit compares consecutive fields of the interlaced video signal having the same parity (i.e., two even fields or two odd fields) and generates a comparison value for each comparison. If a repeating 5 field sequence of 1 small comparison value followed by 4 consecutive large comparison values is detected, the 3:2 pull-down frame motion detector circuit generates a signal which indicates that the interlaced video signal was generated using 3:2 pull-down. Another signal is generated which indicates the 3:2 pull-down sequence.
  • the 3:2 pull-down frame motion detector circuit generates a signal which indicates that the interlaced video signal was not generated using 3:2 pull-down.
  • Fields Al, A2 and A3 were generated from film frame A; fields Bl and B2 were generated from film frame B; fields CI, C2, and C3 were generated from film frame C; fields Dl and D2 were generated from film frame D, and so on.
  • the 3:2 pull-down frame motion detector circuit compares consecutive fields of the interlaced video signal having the same parity yielding a repeating pattern of 1 small comparison value followed by 4 consecutive large comparison values:
  • the signal that indicates whether or not the interlaced video signal was generated using 3:2 pull-down is then provided to a deinterlacer circuit along with the interlaced video signal. If the indication signal indicates that the interlaced video signal was generated using 3:2 pull-down, the deinterlacer circuit uses the merging technique to convert the interlaced video signal into a progressive video signal. On the other hand, if the indication signal indicates that the interlaced video signal was not generated using 3:2 pull-down, the deinterlacer circuit typically uses a motion adaptive technique to convert the interlaced video signal into a progressive video signal.
  • Interlaced video signals that have been generated using 3:2 pull-down are often edited, for example, to insert a television commercial, to cut a scene out of a motion picture, to overlay computer graphics, or to overlay sub-titles, such as credits. Edits that do not occur on a film frame line result in what is commonly referred to as a "bad edit.”
  • Some conventional video deinterlacers do not include the ability to detect bad edits. As a result, it can take up to 5 fields to detect the bad edit. This is illustrated by the following example. Suppose that an interlaced video signal has the following field sequence:
  • the interlaced video signal is edited such that it includes a bad edit resulting in the following sequence:
  • the bad edit occurs between fields E3 andG3.
  • fields FI, F2, Gl, and G2 have been edited out of the interlaced video signal.
  • a conventional video deinterlacer circuit that does not have the ability to detect a bad edit compares consecutive fields of the interlaced video signal having the same parity yielding the following pattern of comparison values:
  • A1 - A3 CV1 (which is a small difference)
  • A2 - B1 CV2 (which is a large difference)
  • A3 - B2 CV3 (which is a large difference)
  • Bl - Cl CV4 (which is a large difference)
  • B2 - C2 CV5 (which is a large difference)
  • C1 - C3 CN6 (which is a small difference)
  • C2 - D1 CV7 (which is a large difference)
  • C3 - D2 CV8 (which is a large difference)
  • Dl - El CV9 (which is a large difference)
  • D2 - E2 CV10 (which is a large difference)
  • E1 - E3 CV11 (which is a small difference)
  • E2 - G3 CV12 (which is a large difference)
  • E3 - H1 CV13 (which is a large difference)
  • G3 - H2 CV14 (which is a large difference)
  • This sequence of comparison values has an initial 10 comparison values which consist of 2 repeating sets of 5 field sequences of 1 small comparison value followed by 4 consecutive large comparison values. However, after this initial 10 comparison values, the sequence of comparison values ceases to have a 5 field sequence of 1 small comparison value followed by 4 consecutive large comparison values. Rather, the sequence now has 1 small comparison value followed by 5 consecutive large comparison values since it includes a bad edit.
  • conventional video deinterlacers cannot detect the bad edit until comparison value CV16 is generated. Since the conventional 3:2 pull-down motion detector circuit initially detects 3:2 pulldown, the video deinterlacer circuit continues to use the merging technique when generating the progressive video signal. After comparison value CV11, this causes fields from different points in time to be merged together when generating the progressive video signal and thus ultimately results in an image that includes highly objectionable artifacts (i.e., feathering or combing).
  • Some conventional video deinterlacers include bad edit detection circuits that can detect bad edits earlier using "look-ahead” techniques.
  • look-ahead techniques A disadvantage with such look-ahead techniques is that they require up to six fields to be simultaneously stored in memory to detect bad edits. This requires a large amount of memory, which is expensive from both an implementation and a production standpoint and thus is undesirable.
  • Another technique used to detect 3:2 pull-down is to use a 3:2 pull-down field motion detector circuit.
  • the 3:2 pull-down field motion detector circuit compares consecutive fields of the interlaced video signal having different parities yielding the following pattern:
  • A3 - Bl CN3 (which is a large difference)
  • Bl - B2 CV4 (which is a small difference)
  • the 3:2 pull-down field motion detector circuit detects the following repeating 5 field sequence: small difference, small difference, large difference, small difference, large difference. Bad edits can be fairly reliably detected using a 3:2 pull-down field motion detector circuit. However, 3:2 pull-down field motion detector circuits are problematic since often times they do not correctly detect when an interlaced video signal was generated using 3:2 pull-down.
  • the present invention provides improved 2:2 pull-down and 3:2 pull-down detection techniques.
  • a 2:2 pull-down field motion detector circuit receives an interlaced video signal and detects whether the interlaced video signal includes a 2:2 pull-down field pattern.
  • the 2:2 pull-down field motion detector circuit accomplishes this by generating a comparison value for consecutive fields of an interlaced video signal having different parities, and detecting a sequence of 1 small comparison value followed by 1 large comparison value. Another signal is generated which indicates the sequence detected by the 2:2 pull-down field motion detector.
  • a 2:2 pull-down frame motion detector circuit receives the interlaced video signal and detects whether the interlaced video signal includes a 2:2 pull-down frame pattern.
  • the 2:2 pull-down frame motion detector circuit accomplishes this by generating a comparison value for consecutive fields of the interlaced video signal having the same parity, generating a difference value for consecutive comparison values, and detecting a sequence of 1 small difference value followed by 1 large difference value.
  • the 2:2 pull-down frame motion detector circuit detects pairs of similar frame motion comparison values. Another signal is generated which indicates the sequence detected by the 2:2 pull-down frame motion detector.
  • the 2:2 pull-down field motion detector circuit does not detect a 2:2 pull-down field pattern or the 2:2 pull-down frame motion detector circuit does not detect a 2:2 pull-down frame pattern, or the 2:2 pull-down sequences detected by the field and frame motion detectors are not the same, a signal is generated that indicates that the interlaced video signal was not generated using 2:2 pull-down.
  • a 3:2 pull-down frame motion detector circuit generates a comparison value for consecutive fields of an interlaced video signal having the same parity, and detects a sequence of 1 small comparison value followed by 2 pairs of similar large comparison values.
  • the 3:2 pull-down frame motion detector circuit If a repeating 5 field sequence of 1 small comparison value followed by 2 pairs of similar large comparison values is detected, the 3:2 pull-down frame motion detector circuit generates a signal which indicates that the interlaced video signal was generated using 3:2 pull-down. Another signal is generated which indicates the sequence detected by the 3:2 pull-down frame motion detector. If a repeating 5 field sequence of 1 small comparison value followed by 2 pairs of similar large comparison values is not detected, the 3:2 pull-down frame motion detector circuit generates a signal which indicates that the interlaced video signal was not generated using 3:2 pull-down.
  • FIG. 1 is a diagram illustrating the 2:2 pull-down technique.
  • FIG. 2 is a diagram illustrating the 3:2 pull-down technique.
  • FIG. 3 is a block diagram of a video deinterlacer system that uses an improved 2:2 pull-down detection technique, according to some embodiments of the present invention.
  • FIG. 4 is a block diagram of a video deinterlacer system that uses an improved 3:2 pull-down detection technique, according to some embodiments of the present invention.
  • FIG. 3 is a block diagram of a video deinterlacer system 300, according to some embodiments of the present invention.
  • Video deinterlacer system 300 includes a 2:2 pull-down detection circuit 310 and a deinterlacer circuit 350.
  • 2:2 pull-down detection circuit 310 includes a field motion detector circuit 320, a frame motion detector circuit 330, and a control circuit 340.
  • the functions performed by video deinterlacer system 300 can be implemented using hardware, firmware/microcode, software, or any combination thereof.
  • Video deinterlacer system 300 can also be implemented on a single integrated circuit device or on multiple integrated circuit devices.
  • 2:2 pull-down field motion detector circuit 320 which can be any conventional 2:2 pull-down field motion detector circuit, receives interlaced video signal IVS and generates a signal (which is output on signal line 322) that indicates whether or not a 2:2 pull-down field pattern has been detected. If a 2:2 pull-down field pattern has been detected, another signal (which is output on signal line 324) is generated which indicates the 2:2 sequence.
  • the 2:2 sequence signal on signal line 324 is a timing signal that allows deinterlacer circuit 350 to properly merge the fields of interlaced video signal IVS (e.g., so that fields Al and A2 will be merged rather than fields A2 and Bl).
  • 2:2 pull-down field motion detector circuit 320 compares consecutive fields of interlaced video signal IVS having different parities and generates a comparison value for each comparison. If a 2:2 pull-down field pattern is detected by 2:2 pull-down field motion detector circuit, then 2:2 pull-down field motion detector circuit generates and outputs a signal on signal line 322 that indicates that a 2:2 pull-down pattern has been detected, and generates and outputs a signal on signal line 324 which indicates the 2:2 sequence.
  • a 2:2 pull-down field pattern is a repeating sequence of 1 small comparison value followed by 1 large comparison value.
  • 2:2 pull-down field motion detector circuit determines whether a 2:2 pull-down field pattern is detected by 2:2 pull-down field motion detector circuit. If a 2:2 pull-down field pattern is not detected by 2:2 pull-down field motion detector circuit, then 2:2 pull-down field motion detector circuit generates and outputs a signal on signal line 322 that indicates that a 2:2 pull-down pattern has not been detected. Since the operation of conventional 2:2 pull-down field detector circuits are well known, the operation of 2:2 pull-down field detector circuit 320 is not described any further herein.
  • 2:2 pull-down field motion detector circuit 320 one problem with conventional 2:2 pull-down field motion detector circuits, such as 2:2 pull-down field motion detector circuit 320 is that they often falsely detect a 2:2 pull-down field pattern.
  • the merging technique is mistakenly used to generate the progressive video signal from the interlaced video signal that ultimately results in a displayed image that includes highly objectionable artifacts (i.e., feathering or combing).
  • a frame motion detector circuit such as 2:2 pull-down frame motion detector circuit 330, in addition to a 2:2 pull-down field motion detector circuit.
  • 2:2 pull-down frame motion detector circuit 330 receives interlaced video signal IVS and generates a signal (which is output on signal line 332) that indicates whether or not a 2:2 pull-down frame pattern has been detected. If a 2:2 pull-down frame pattern has been detected, another signal (which is output on signal line 334) is generated which indicates the 2:2 sequence.
  • the 2:2 sequence signal on signal line 334 is a timing signal that allows deinterlacer circuit 350 to properly merge the fields of interlaced video signal IVS (e.g., so that fields Al and A2 will be merged rather than fields A2 and Bl).
  • frame motion detector circuit 330 compares consecutive fields having the same parity and generates a comparison value for each comparison.
  • Consecutive comparison values are then compared to generate difference values. If a 2:2 pull-down frame pattern is detected by frame motion detector circuit 330, then frame motion detector circuit 330 generates and outputs a signal on signal line 332 that indicates that interlaced video signal IVS was generated using 2:2 pulldown, and generates and outputs a signal on signal line 334 which indicates the 2:2 sequence.
  • a 2:2 pull-down frame pattern is a repeating sequence of 1 small difference value followed by 1 large difference value.
  • frame motion detector circuit 330 determines whether a 2:2 pull-down frame pattern is detected by frame motion detector circuit 330. If a 2:2 pull-down frame pattern is not detected by frame motion detector circuit 330, then frame motion detector circuit 330 generates and outputs a signal on signal line 332 that indicates that interlaced video signal IVS was not generated using 2:2 pull-down.
  • 2:2 pull-down frame motion detector circuit 330 compares consecutive fields having the same parity to generate comparison values. Thus, 2:2 pull-down frame motion detector circuit 330 compares fields Al and Bl. This comparison results in a first comparison value CVl. 2:2 pull-down frame motion detector circuit 330 then compares fields A2 and B2. This comparison results in a second comparison value CV2. 2:2 pull-down field motion detector circuit 330 then compares fields Bl and CI. This comparison results in a third comparison value CV3. 2:2 pull-down frame motion detector circuit 330 then compares fields B2 and C2. This results in a fourth comparison value CV4. 2:2 pull-down frame motion detector circuit 330 then compares fields CI and Dl. This results in a fifth comparison value CV5. 2:2 pulldown frame motion detector circuit 330 then compares fields C2 and D2. This results in a sixth comparison value CV6. This same process is performed as frame motion detector circuit 330 receives each new field. The comparisons and the resulting comparison values are as follows:
  • 2:2 pull-down frame motion detector circuit 330 compares the recently generated comparison value with the previously generated comparison value to generate a difference value.
  • 2:2 pull-down frame motion detector circuit 330 compares comparison values CN1 and CV2.
  • This comparison results in a first difference value DV1, which is a small difference since comparison values CVl and CV2 were each generated by comparing fields that were generated from the same two film frames.
  • 2:2 pull-down frame motion detector circuit 330 compares comparison values CV2 and CV3.
  • This comparison results in a second difference value DV2, which is a large difference since comparison values CV2 and CV3 were not generated by comparing fields that were generated from the same two film frames.
  • 2:2 pull-down frame motion detector circuit 330 compares comparison values CV3 and CV4. This comparison results in a third difference value DV3, which is a small difference since comparison values CV3 and CV4 were each generated by comparing fields that were generated from the same two film frames. 2:2 pull-down frame motion detector circuit 330 then compares comparison values CV4 and CV5. This comparison results in a fourth difference value DV4, which is a large difference since comparison values CV4 and CV5 were not generated by comparing fields that were generated from the same two film frames. 2:2 pull-down frame motion detector circuit 330 then compares comparison values CV5 and CV6. This comparison results in a fifth difference value DV5, which is a small difference since comparison values CV5 and CV6 were each generated by comparing fields that were generated from the same two film frames. The comparisons and the resulting difference values are as follows:
  • 2:2 pull-down frame motion detector circuit 330 compares consecutive fields having the same parity to generate comparison values.
  • the comparisons and the resulting comparison values are as follows:
  • 2:2 pull-down frame motion detector circuit 330 compares the recently generated comparison value with the previously generated comparison value to generate a difference value.
  • the comparisons and the resulting difference values are as follows:
  • CV2 - CV3 DV2 (which is a large difference)
  • CV3 - CV4 DV3 (which is a large difference)
  • comparison values (from which each difference value is generated) were not generated by comparing fields that were generated from the same two film frames, all of the difference values in this sequence are large difference values. Note that the difference values may be a random pattern of small or large difference values. Thus, a 2:2 pull-down frame pattern does not emerge, and 2:2 pull-down frame motion detector circuit 330 generates and outputs a signal on signal line 332 that indicates that interlaced video signal IVS was not generated using 2:2 pull-down.
  • Control circuit 340 receives two different pairs of signals.
  • the first pair of signals includes the signal that indicates whether interlaced video signal IVS was generated using 2:2 pull-down on signal line 322 and the signal that indicates the 2:2 sequences on signal line 324.
  • the second pair of signals includes the signal that indicates whether interlaced video signal IVS was generated using 2:2 pull-down on signal line 332 and the signal that indicates the 2:2 sequences on signal line 334.
  • Control circuit 340 compares the signals that indicate whether interlaced video signal IVS was generated using 2:2 pull-down (i.e., the signals on signal lines 322 and 332) and if they both indicate that interlaced video signal IVS was generated using 2:2 pull-down, control circuit 340 generates and outputs a signal on signal line 342 that causes deinterlacer circuit 350 to generate progressive video signal PVS from interlaced video signal IVS using the merging technique. Control circuit 340 also compares the signals that indicate the 2:2 sequence (i.e., the signals on signal lines 324 and 334) and if they both indicate the same 2:2 sequence, control circuit 340 generates and output a signal on signal line 344 that indicates the 2:2 sequence.
  • the 2:2 sequence signal on signal line 344 is a timing signal that allows deinterlacer circuit 350 to properly merge the fields of interlaced video signal IVS (e.g., so that fields Al and A2 will be merged rather than fields A2 and Bl).
  • control circuit 340 If, on the other hand, the two signals that indicate whether interlaced video signal IVS was generated using 2:2 pull-down (i.e., the signals on signal lines 322 and 332) do not both indicate that interlaced video signal IVS was generated using 2:2 pull-down, control circuit 340 generates and outputs a signal on signal line 342 that causes deinterlacer circuit 350 to generate progressive video signal PNS from interlaced video signal IVS using a motion adaptive technique, a motion-compensated technique, and/or an interpolation technique.
  • control circuit 340 can generate and output a signal on signal line 342 that causes deinterlacer circuit 350 to generate progressive video signal PVS from interlaced video signal IVS using a motion adaptive technique, a motion-compensated technique, and/or an interpolation technique.
  • control circuit 340 can be omitted and signal lines 322, 324, 332, and 334 can be connected directly to deinterlacer circuit 350.
  • Deinterlacer circuit 350 can then perform the same functions that control circuit 340 performs as described above.
  • the 2:2 frame motion detector circuit detects whether every other difference value is a small difference value. If every other difference value is a small difference value, 2:2 pull-down frame motion detector circuit 330 detects this and generates and outputs a signal that indicates that interlaced video signal IVS was generated using 2:2 pull-down and a signal that indicates the 2:2 sequence. In this embodiment, the other difference values are treated as "don't cares.”
  • the technique employed by 2:2 pull-down frame motion detector circuit of this alternative embodiment is illustrated by the following example. Suppose the interlaced video signal IVS was generated using 2:2 pull-down and has the following field sequence:
  • Comparison values are generated from the field sequences as described above, and then the following difference values are generated from the comparison values.
  • CV4 - CV5 DV4 (which is a don't care)
  • the 2:2 frame motion detector circuit detects that every other difference value is a small difference and thus detects that the interlaced video signal was generated using 2:2 pull-down.
  • the 2:2 frame motion detector circuit initially detects a repeating sequence of 1 small difference value followed by 1 large difference value. After this initial sequence, which has a programmable duration, is detected, the 2:2 frame motion detector circuit then detects whether every other difference value is a small difference value. This detection process is similar to the detection process used by some 2:2 field motion detector circuits, which initially detect a repeating sequence of 1 small comparison value followed by 1 large comparison value. After this initial sequence is detected, the 2:2 field motion detector circuit then detects whether every other comparison value is a small comparison value.
  • conventional 3:2 pull-down frame motion detector circuits compare the difference between fields of the same parity and detect a repeating 5 field sequence of 1 small comparison value followed by 4 consecutive large comparison values to determine whether the interlaced video signal was generated using 3:2 pulldown.
  • some conventional 3:2 pull-down frame motion detector circuits use look-ahead techniques, which require up to 6 fields to be stored in memory. This requires a large amount of memory, which is expensive from both an implementation and production standpoint. Applicant has discovered a technique that can be used to detect bad edits in interlaced video signals generated using 3:2 pulldown that only requires an economical 2 fields to be stored in memory at any given time.
  • FIG. 4 is a block diagram of a video deinterlacer system 400, according to some embodiments of the present invention.
  • Video deinterlacer system 400 includes a 3:2 pull-down detection circuit 410 and a deinterlacer circuit 450.
  • 3:2 pull-down detection circuit 410 includes a frame motion detector circuit 430.
  • the functions performed by video deinterlacer system 400 can be implemented using hardware, firmware/microcode, software, or any combination thereof.
  • Video deinterlacer system 400 can also be implemented on a single integrated circuit device or on multiple integrated circuit devices.
  • 3:2 pull-down frame motion detector circuit 430 receives interlaced video signal IVS and generates a signal that indicates whether or not a 3:2 pull-down frame pattern has been detected.
  • frame motion detector circuit 430 compares consecutive fields having the same parity and generates a comparison value for each comparison. Consecutive comparison values are analyzed to determine whether a repeating 5 field sequence of 1 small comparison value followed by 2 pairs of similar large comparison values exists. If this pattern is detected by frame motion detector circuit 430, then frame motion detector circuit 430 generates and outputs a signal (on signal line 442) that indicates that interlaced video signal IVS was generated using 3:2 pull-down, and generates and outputs a signal (on signal line 444) that indicates the 3:2 sequence.
  • the 3:2 sequence signal on signal line 444 is a timing signal that allows deinterlacer circuit 450 to properly merge the fields of interlaced video signal IVS (e.g., so that fields A2 and A3 will be merged rather than fields A3 and Bl).
  • frame motion detector circuit 430 if this pattern is not detected by frame motion detector circuit 430, then frame motion detector circuit 430 generates and outputs a signal (on signal line 442) that indicates that interlaced video signal IVS was not generated using 3:2 pull-down.
  • Interlaced video signal IVS has the following field sequence:
  • 3:2 pull-down frame motion detector circuit 430 compares consecutive fields having the same parity to generate comparison values. Thus, 3:2 pull-down frame motion detector circuit 430 compares fields Al and A3. This comparison results in a first comparison value CVl, which is a small difference since fields Al and A3 were generated from the same film frame. 3:2 pull-down frame motion detector circuit 430 then compares fields A2 and Bl. This comparison results in a second comparison value CV2, which is a large difference since fields A2 and Bl were generated from different film frames. 3:2 pull-down field motion detector circuit 430 then compares fields A3 and B2. This comparison results in a third comparison value CV3, which is a large difference since fields A3 and B2 were generated from different film frames.
  • comparison value CV3 is similar to comparison value CV2 since they are both generated by comparing fields that were generated from the same two film frames.
  • comparison values CV2 and CV3 make up a first pair of similar large comparison values.
  • 3:2 pull-down frame motion detector circuit 430 compares fields Bl and CI. This results in a fourth comparison value CV4, which is a large difference since fields Bl and CI were generated from different film frames.
  • 3:2 pull-down frame motion detector circuit 430 compares fields B2 and C2. This results in a fifth comparison value CV5, which is a large difference since fields B2 and C2 were generated from different film frames.
  • comparison value CV5 is similar to comparison value CV4 since they are both generated by comparing fields that were generated from the same two film frames. Thus, comparison values CV4 and CV5 make up a second pair of similar large comparison values. This same process is performed as frame motion detector circuit 430 receives each new field. The comparisons and the resulting comparison values are as follows:
  • A3 - B2 CV3 (which is a large difference and is similar to CV2)
  • B 1 - C 1 CV4 (which is a large difference)
  • C2 - D1 CV7 (which is a large difference)
  • C3 - D2 CV8 (which is a large difference and is similar to CV7)
  • Dl - El CV9 (which is a large difference)
  • 3:2 pull-down frame motion detector circuit 430 detects a repeating 5 field sequence of 1 small comparison value followed by 2 pairs of similar large comparison values. If a repeating 5 field sequence of 1 small comparison value followed by 2 pairs of similar large comparison values is detected, 3:2 pull-down frame motion detector circuit 430 generates and outputs a signal on signal line 442 which indicates that the interlaced video signal was generated using 3:2 pull-down, and generates and outputs a signal on signal line 444 that indicates the 3:2 sequence.
  • the 3:2 pull-down frame motion detector circuit 430 If a repeating 5 field sequence of 1 small comparison value followed by 2 pairs of similar large comparison values is not detected, the 3:2 pull-down frame motion detector circuit 430 generates and outputs a signal on signal line 442 which indicates that the interlaced video signal was not generated using 3:2 pull-down.
  • Interlaced video signal IVS was generated using 3:2 pull-down and includes a bad edit.
  • Interlaced video signal IVS has the following sequence:
  • Al - A3 CVl (which is a small difference)
  • A2 - Bl CV2 (which is a large difference)
  • A3 - B2 CV3 (which is a large difference and is similar to CV2)
  • Bl - CI CV4 (which is a large difference)
  • B2 - C2 CV5 (which is a large difference and is similar to CV4)
  • CI - C3 CV6 (which is a small difference)
  • C2 - Dl CV7 (which is a large difference)
  • C3 - D2 CV8 (which is a large difference and is similar to CV7)
  • 3:2 pull-down frame motion detector circuit 430 quickly detects a bad edit since a small comparison value (i.e., comparison value CVl 1) is followed by two dissimilar large comparison values (i.e., comparison value CV12 and CV13) rather than a pair of similar comparison values. It is important to note that comparison value CVl 3 is not similar to comparison value CVl 2 since they are not generated by comparing fields that were generated from the same two film frames. Since 3:2 pulldown frame motion detector circuit 430 does not detect a repeating 5 field sequence of 1 small comparison value followed by 2 pairs of similar large comparison values, 3:2 pull-down frame motion detector 430 generates and outputs a signal on signal line 442 that indicates that a bad edit has occurred. This signal is then output to deinterlacer circuit 450.
  • a small comparison value i.e., comparison value CVl 1
  • CV12 and CV13 two dissimilar large comparison values
  • 3:2 pull-down frame motion detector circuit 430 compares consecutive fields having the same parity to generate comparison values.
  • the comparisons and the resulting comparison values are as follows:
  • a - C CVl (which is either a small or a large difference)
  • B - D CV2 (which is a large difference and is not similar to CVl)
  • 3:2 pull-down frame motion detector circuit 430 quickly determines that the interlaced video signal was not generated using 3:2 pull-down since comparison values CV2 and CV3 are dissimilar. Since 3:2 pull-down frame motion detector circuit 430 does not detect a 5 field sequence of 1 small comparison value followed by 2 pairs of similar large comparison values, 3:2 pull-down frame motion detector 430 generates a signal that indicates that the interlaced video signal was not generated using 3:2 pull-down. 3:2 pull-down frame motion detector 430 continues to generate the signal that indicates that the interlaced video signal was not generated using 3:2 pull-down, as long as the received field pattern is random.
  • deinterlacer circuit 450 If the signal generated by frame motion detector 430 (on signal line 442) indicates that the interlaced video signal INS was generated using 3:2 pull-down, deinterlacer circuit 450 generates progressive video signal PVS from interlaced video signal IVS using the merging technique. If, on the other hand, the signal generated by frame motion detector 430 (on signal line 442) indicates that the interlaced video signal IVS was not generated using 3:2 pull-down, deinterlacer circuit 450 generates progressive video signal PVS from interlaced video signal IVS using a motion adaptive technique, a motion-compensated technique, or an interpolation technique.
  • frame motion detector system 400 can quickly detect bad edits while only needing to store an economical 2 fields of video.
  • the 3:2 frame motion detector circuit initially detects a repeating 5 field sequence of 1 small comparison value followed by 2 pairs of similar large comparison values. After this initial sequence, which has a programmable duration, is detected, the 3:2 frame motion detector circuit then detects a repeating 5 field sequence of 1 small comparison value followed by 2 pairs of similar comparison values. These pairs of comparison values may either have large or small values.
  • This embodiment is advantageous in situations where an interlaced video signal is generated using 3:2 pull-down, but has very little motion from frame-to-frame. In these situations, it is desirable to continue to use the merging technique to generate the progressive video signal.
  • A3 - B2 CV3 (which is a large difference and is similar to CV2)
  • E3 - F2 CVl 3 (which is a large or a small difference and is similar to CVl 2)
  • FI - Gl CVl 4 (which is a large or a small difference)
  • a second and/or third 3:2 pull-down frame motion detector circuit can be used.
  • the second 3:2 frame motion detector circuit detects a repeating 5 field sequence of 1 small comparison value followed by a pair of dissimilar large comparison values and then followed by a pair of similar large comparison values to determine whether the interlaced video signal was generated using 3:2 pull-down.
  • the third 3:2 frame motion detector circuit detects a repeating 5 field sequence of 1 small comparison value followed by a pair of similar large comparison values and then followed by dissimilar large comparison values to determine whether the interlaced video signal was generated using 3:2 pull-down.
  • A3 - B2 CV3 (which is a large difference and is not similar to CV2)
  • B 1 - C 1 CV4 (which is a large difference)
  • E3 - F2 CVl 3 (which is a large difference and is not similar to CVl 2)
  • FI - Gl CVl 4 (which is a large difference)
  • F2 - G2 CVl 5 (which is a large difference and is similar to CVl 4)
  • Gl - G3 CVl 6 (which is a small difference)
  • A3 - B2 CV3 (which is a large difference and is similar to CV2)
  • E2 - FI CVl 2 (which is a large difference)
  • E3 - F2 CVl 3 (which is a large difference and is similar to CVl 2)
  • FI - Gl CV14 (which is a large difference)
  • the 3:2 pull-down field motion detector circuit compares consecutive fields of the interlaced video signal having different parities and detects a repeating 5 field sequence of a pair of small comparison values, followed by a "don't care” comparison value, followed by a small comparison value, followed by a "don't care” comparison value.
  • the 3:2 frame motion detector then generates comparison values as follows:
  • A2 - Bl CV2 (which is a large difference)
  • A3 - B2 CV3 (which is a large difference and is similar to CV2)
  • the 3:2 field motion detector must then generate comparisons values as follows:
  • A2 - A3 CV2 (which is a small difference)
  • A3 - B 1 CV3 (which is a don't care)
  • E2 - E3 CVl 2 (which is a small difference)
  • E3 - FI CVl 3 (which is a don't care)
  • the improved 2:2 pull-down and 3:2 pull-down detection techniques of the present invention can be utilized in a wide variety of video signal processing applications.
  • the 2:2 pull-down detection technique and/or the 3:2 pull-down detection technique of the present invention can be implemented in a single integrated circuit device that is used to convert an interlaced video signal into a progressive video signal.
  • the integrated circuit device can be integrated into a digital video disc (DVD) player, a DVD recorder, a computer graphics controller, or an LCD panel controller.
  • DVD digital video disc
  • the 2:2 pull-down detection technique and/or the 3:2 pull-down detection technique of the present invention can be implemented in the firmware of a computer processor, such as a CPU, a DSP, or a media processor, that is used to convert an interlaced video signal into a progressive video signal.
  • the 2:2 pull-down detection technique and/or the 3:2 pull-down detection technique of the present invention can also be used in a wide variety of digital video compression applications, a wide variety of digital video transmission applications, in progressive- scan TVs, in high definition televisions (HDTVs), and in LCD panels.
  • the 2:2 pull-down detection technique and/or the 3:2 pull-down detection technique of the present invention can be implemented in software for such applications as software based DVD decoders used in computers and/or video game consoles.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Television Systems (AREA)
PCT/US2003/010858 2002-04-09 2003-04-08 2:2 and 3:2 pull-down detection techniques Ceased WO2003088641A2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP03746672A EP1510069A4 (en) 2002-04-09 2003-04-08 PULLDOWN DETECTION TECHNIQUES 2: 2 AND 3: 2
JP2003585418A JP4533632B2 (ja) 2002-04-09 2003-04-08 ２：２プルダウン検出技法

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US10/119,999 2002-04-09
US10/119,999 US7202907B2 (en) 2002-04-09 2002-04-09 2:2 and 3:2 pull-down detection techniques

Publications (2)

Publication Number Publication Date
WO2003088641A2 true WO2003088641A2 (en) 2003-10-23
WO2003088641A3 WO2003088641A3 (en) 2004-12-23

Family

ID=28674624

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2003/010858 Ceased WO2003088641A2 (en) 2002-04-09 2003-04-08 2:2 and 3:2 pull-down detection techniques

Country Status (4)

Country Link
US (2) US7202907B2 (enExample)
EP (1) EP1510069A4 (enExample)
JP (1) JP4533632B2 (enExample)
WO (1) WO2003088641A2 (enExample)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005318611A (ja) * 2004-04-30 2005-11-10 Matsushita Electric Ind Co Ltd ビデオシーケンスにおけるフィルムモード検出方法、フィルムモード検出器、動き補償方法及び動き補償器
GB2420040A (en) * 2004-11-05 2006-05-10 Hitachi Ltd Image correction using additional information input with image signal
JP2006237695A (ja) * 2005-02-22 2006-09-07 Nec Electronics Corp プルダウン検出装置及びプルダウン検出方法
JP2006237696A (ja) * 2005-02-22 2006-09-07 Nec Electronics Corp プルダウン検出装置及びプルダウン検出方法
JP2006270912A (ja) * 2005-02-22 2006-10-05 Nec Electronics Corp プルダウン検出装置及びプルダウン検出方法

Families Citing this family (43)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110013081A1 (en) * 2001-01-11 2011-01-20 Pixelworks, Inc. System and method for detecting a non-video source in video signals
CA2330854A1 (en) * 2001-01-11 2002-07-11 Jaldi Semiconductor Corp. A system and method for detecting a non-video source in video signals
US7202907B2 (en) 2002-04-09 2007-04-10 Zoran Corporation 2:2 and 3:2 pull-down detection techniques
KR100854091B1 (ko) * 2002-07-13 2008-08-25 삼성전자주식회사 영상신호의 필름 모드 검출장치 및 방법
JP4017498B2 (ja) * 2002-11-05 2007-12-05 松下電器産業株式会社 撮像装置
KR100927143B1 (ko) 2002-11-27 2009-11-18 삼성전자주식회사 움직임 검출장치 및 방법
KR20040054032A (ko) * 2002-12-16 2004-06-25 삼성전자주식회사 영상신호의 포맷 검출장치 및 방법
KR20040055059A (ko) * 2002-12-20 2004-06-26 삼성전자주식회사 영상포맷의 변환장치 및 방법
US7154555B2 (en) * 2003-01-10 2006-12-26 Realnetworks, Inc. Automatic deinterlacing and inverse telecine
JP2005026885A (ja) * 2003-06-30 2005-01-27 Toshiba Corp テレビジョン受信装置及びその制御方法
KR20050011069A (ko) * 2003-07-21 2005-01-29 삼성전자주식회사 배드에딧으로 인한 콤을 제거할 수 있는 영상신호판별장치 및 그 판별방법
KR20050011068A (ko) * 2003-07-21 2005-01-29 삼성전자주식회사 필름모드 판별장치 및 그 방법
TWI225365B (en) * 2003-12-12 2004-12-11 Sunplus Technology Co Ltd Detecting apparatus capable of detecting bad edition of image signal
US8861589B2 (en) * 2004-01-30 2014-10-14 Broadcom Corporation Detection and phase lock of pull-down video
US7557861B2 (en) * 2004-01-30 2009-07-07 Broadcom Corporation Reverse pull-down video using corrective techniques
CN100372359C (zh) * 2004-02-04 2008-02-27 凌阳科技股份有限公司 可侦测影像讯号不良编辑的侦测装置
TWI240558B (en) * 2004-03-22 2005-09-21 Realtek Semiconductor Corp Method and apparatus for determining video signals
TWI234388B (en) * 2004-04-23 2005-06-11 Himax Tech Inc De-interlacing device having pattern recognition unit and method thereof
KR100633143B1 (ko) * 2004-08-26 2006-10-11 삼성전자주식회사 배드에딧 검출이 가능한 필름모드 판별장치 및 그 방법
US7468756B2 (en) * 2004-10-05 2008-12-23 Broadcom Corporation Detection and phase lock of 2:2 and 3:2 pull-down video
US7468757B2 (en) * 2004-10-05 2008-12-23 Broadcom Corporation Detection and correction of irregularities while performing inverse telecine deinterlacing of video
US7349029B1 (en) * 2005-01-19 2008-03-25 Kolorific, Inc. Method and apparatus for de-interlacing interlaced video fields originating from a progressive video source
JP3916637B2 (ja) * 2005-03-08 2007-05-16 三菱電機株式会社 映像信号処理装置、映像信号処理方法、及び映像信号表示装置
US7868948B2 (en) * 2005-09-12 2011-01-11 Sony Corporation Mage signal processing apparatus, image signal processing method and program for converting an interlaced signal into a progressive signal
JP2007129400A (ja) * 2005-11-02 2007-05-24 Sharp Corp フィルムモード検出装置及び映像表示装置
US7705913B2 (en) * 2005-12-20 2010-04-27 Lsi Corporation Unified approach to film mode detection
KR100722773B1 (ko) * 2006-02-28 2007-05-30 삼성전자주식회사 동영상에서 그래픽 영역을 검출하는 방법 및 장치
US8035748B2 (en) * 2006-07-12 2011-10-11 Broadcom Corporation Method and system for reducing composite video cross-chroma artifacts in movie material for a deinterlacer
JP2008035436A (ja) * 2006-07-31 2008-02-14 Fujitsu Ltd プルダウンシーケンス検出装置およびプルダウンシーケンス検出方法
US8619190B2 (en) * 2006-09-07 2013-12-31 Texas Instruments Incorporated Film mode detection
US7808552B1 (en) * 2006-10-25 2010-10-05 Vimicro Corporation Techniques for detecting a pulldown field sequence
JP4749314B2 (ja) * 2006-11-28 2011-08-17 富士通株式会社 インタレース／プログレッシブ変換方法および変換装置
US8957961B2 (en) * 2006-12-27 2015-02-17 Intel Corporation Method and sytem for telecine detection and restoration
KR101197149B1 (ko) * 2007-07-10 2012-11-08 삼성전자주식회사 디스플레이장치 및 그 제어방법
US8891011B2 (en) 2007-08-23 2014-11-18 Qualcomm Incorporated Systems and methods for combining deinterlacing and frame rate decimation for video format conversion
JP4935632B2 (ja) * 2007-11-07 2012-05-23 ソニー株式会社 画像処理装置、画像処理方法および画像処理プログラム
JP5123643B2 (ja) * 2007-11-12 2013-01-23 ルネサスエレクトロニクス株式会社 映像処理装置
US20090180544A1 (en) * 2008-01-11 2009-07-16 Zoran Corporation Decoding stage motion detection for video signal deinterlacing
US8593572B2 (en) * 2008-01-30 2013-11-26 Csr Technology Inc. Video signal motion detection
JP4388126B1 (ja) * 2008-06-27 2009-12-24 株式会社東芝 プルダウン信号検出装置及びプルダウン信号検出方法、順次走査変換装置
US20110001873A1 (en) * 2008-08-06 2011-01-06 Daniel Doswald Frame rate converter for input frames with video and film content
US9491398B1 (en) 2010-12-21 2016-11-08 Pixelworks, Inc. System and method for processing assorted video signals
JPWO2012137269A1 (ja) * 2011-04-01 2014-07-28 パナソニック株式会社 映像信号のフレームレート変換装置及びその方法、そのフレームレート変換を用いた映像処理装置

Family Cites Families (44)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4486785A (en) 1982-09-30 1984-12-04 International Business Machines Corporation Enhancement of video images by selective introduction of gray-scale pels
US4876596A (en) 1988-10-25 1989-10-24 Faroudja Y C Film-to-video converter with scan line doubling
US4982280A (en) * 1989-07-18 1991-01-01 Yves C. Faroudja Motion sequence pattern detector for video
JP3398396B2 (ja) * 1992-01-07 2003-04-21 ソニー株式会社 映像信号処理回路
US5291280A (en) 1992-05-05 1994-03-01 Faroudja Y C Motion detection between even and odd fields within 2:1 interlaced television standard
US5398071A (en) 1993-11-02 1995-03-14 Texas Instruments Incorporated Film-to-video format detection for digital television
JP2933487B2 (ja) 1994-07-15 1999-08-16 松下電器産業株式会社 クロマフォーマット変換の方法
US5705628A (en) * 1994-09-20 1998-01-06 Whitehead Institute For Biomedical Research DNA purification and isolation using magnetic particles
US5886745A (en) 1994-12-09 1999-03-23 Matsushita Electric Industrial Co., Ltd. Progressive scanning conversion apparatus
US5563651A (en) * 1994-12-30 1996-10-08 Thomson Consumer Electronics, Inc. Method and apparatus for identifying video fields produced by film sources employing 2-2 and 3-2 pull down sequences
US5684544A (en) 1995-05-12 1997-11-04 Intel Corporation Apparatus and method for upsampling chroma pixels
US6058140A (en) 1995-09-08 2000-05-02 Zapex Technologies, Inc. Method and apparatus for inverse 3:2 pulldown detection using motion estimation information
KR100192270B1 (ko) 1996-02-03 1999-06-15 구자홍 에이치디티브이 비데오 디코더 회로
US5852473A (en) * 1996-02-20 1998-12-22 Tektronix, Inc. 3-2 pulldown detector
US5712687A (en) 1996-04-25 1998-01-27 Tektronix, Inc. Chrominance resampling for color images
JP3495507B2 (ja) * 1996-05-24 2004-02-09 日本放送協会 画像変換タイミング検出装置
JPH10145601A (ja) 1996-11-08 1998-05-29 Ricoh Co Ltd 画像処理装置
US6141056A (en) 1997-08-08 2000-10-31 Sharp Laboratories Of America, Inc. System for conversion of interlaced video to progressive video using horizontal displacement
US6262773B1 (en) 1997-09-15 2001-07-17 Sharp Laboratories Of America, Inc. System for conversion of interlaced video to progressive video using edge correlation
EP1013089B1 (en) 1997-10-06 2005-06-22 Silicon Image, Inc. Digital video system and methods for providing same
US6014182A (en) 1997-10-10 2000-01-11 Faroudja Laboratories, Inc. Film source video detection
US6208350B1 (en) 1997-11-04 2001-03-27 Philips Electronics North America Corporation Methods and apparatus for processing DVD video
US6295091B1 (en) 1997-11-07 2001-09-25 Intel Corporation Method and apparatus for de-interlacing video fields for superior edge preservation
US6239847B1 (en) 1997-12-15 2001-05-29 Netergy Networks, Inc. Two pass multi-dimensional data scaling arrangement and method thereof
US6157412A (en) 1998-03-30 2000-12-05 Sharp Laboratories Of America, Inc. System for identifying video fields generated from film sources
US6501507B1 (en) 1998-05-13 2002-12-31 Barth Alan Canfield Multimode interpolation filter as for a TV receiver
US6463550B1 (en) * 1998-06-04 2002-10-08 Compaq Information Technologies Group, L.P. Computer system implementing fault detection and isolation using unique identification codes stored in non-volatile memory
US6133960A (en) 1998-06-26 2000-10-17 Lsi Logic Corporation Field-based upsampling method for performing zoom in a digital video system
US6587158B1 (en) 1998-07-23 2003-07-01 Dvdo, Inc. Method and apparatus for reducing on-chip memory in vertical video processing
DE69920055T2 (de) * 1999-02-26 2005-09-15 Stmicroelectronics Asia Pacific Pte Ltd. Verfahren und vorrichtung zur bestimmung von zeilensprung-/nichtzeilensprungbildern und erkennung von wiederholten halbbildern und szenenänderungen
US6377313B1 (en) 1999-09-02 2002-04-23 Techwell, Inc. Sharpness enhancement circuit for video signals
US6573940B1 (en) 1999-09-02 2003-06-03 Techwell, Inc Sample rate converters for video signals
US6297801B1 (en) 1999-09-10 2001-10-02 Intel Corporation Edge-adaptive chroma up-conversion
US6563550B1 (en) 2000-03-06 2003-05-13 Teranex, Inc. Detection of progressive frames in a video field sequence
US6680752B1 (en) 2000-03-31 2004-01-20 Ati International Srl Method and apparatus for deinterlacing video
US6647061B1 (en) 2000-06-09 2003-11-11 General Instrument Corporation Video size conversion and transcoding from MPEG-2 to MPEG-4
SE522121C2 (sv) 2000-10-04 2004-01-13 Axis Ab Metod och anordning för digital behandling av frekvent uppdaterade bilder från en kamera
KR100351159B1 (ko) 2000-12-06 2002-09-05 엘지전자 주식회사 영상 복원 장치 및 방법
US7006147B2 (en) 2000-12-22 2006-02-28 Thomson Lincensing Method and system for MPEG chroma de-interlacing
JP2002290927A (ja) * 2001-03-27 2002-10-04 Toshiba Corp フィルムモード判別回路、フィルムモード判別方法、及び順次走査変換テレビジョン受像機
JP3734430B2 (ja) 2001-05-09 2006-01-11 三菱電機株式会社 輪郭補正回路
KR100423504B1 (ko) 2001-09-24 2004-03-18 삼성전자주식회사 영상신호의 라인보간 장치 및 방법
US7202907B2 (en) 2002-04-09 2007-04-10 Zoran Corporation 2:2 and 3:2 pull-down detection techniques
US7136417B2 (en) 2002-07-15 2006-11-14 Scientific-Atlanta, Inc. Chroma conversion optimization

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2005318611A (ja) * 2004-04-30 2005-11-10 Matsushita Electric Ind Co Ltd ビデオシーケンスにおけるフィルムモード検出方法、フィルムモード検出器、動き補償方法及び動き補償器
GB2420040A (en) * 2004-11-05 2006-05-10 Hitachi Ltd Image correction using additional information input with image signal
GB2420040B (en) * 2004-11-05 2006-09-27 Hitachi Ltd Image display apparatus
US7450182B2 (en) 2004-11-05 2008-11-11 Hitachi, Ltd. Image display apparatus and picture quality correction
JP2006237695A (ja) * 2005-02-22 2006-09-07 Nec Electronics Corp プルダウン検出装置及びプルダウン検出方法
JP2006237696A (ja) * 2005-02-22 2006-09-07 Nec Electronics Corp プルダウン検出装置及びプルダウン検出方法
JP2006270912A (ja) * 2005-02-22 2006-10-05 Nec Electronics Corp プルダウン検出装置及びプルダウン検出方法
US7999876B2 (en) 2005-02-22 2011-08-16 Renesas Electronics Corporation Pull-down detection apparatus and pull-down detection method
US8189103B2 (en) 2005-02-22 2012-05-29 Renesas Electronics Corporation Pull-down detection apparatus and pull-down detection method
US8330858B2 (en) 2005-02-22 2012-12-11 Renesas Electronics Corporation Pull-down detection apparatus and pull-down detection method

Also Published As

Publication number Publication date
EP1510069A2 (en) 2005-03-02
WO2003088641A3 (en) 2004-12-23
JP4533632B2 (ja) 2010-09-01
US7202907B2 (en) 2007-04-10
JP2005522951A (ja) 2005-07-28
US20060012707A1 (en) 2006-01-19
US7612829B2 (en) 2009-11-03
US20030189667A1 (en) 2003-10-09
EP1510069A4 (en) 2010-08-04

Similar Documents

Publication Publication Date Title
WO2003088641A2 (en) 2:2 and 3:2 pull-down detection techniques
AU2003248047B2 (en) Method and video processing unit for processing a video signal
JP5008826B2 (ja) 高精細度デインタレース／フレーム倍増回路およびその方法
US7961253B2 (en) Method of processing fields of images and related device for data lines similarity detection
US20080055477A1 (en) Method and System for Motion Compensated Noise Reduction
US7218354B2 (en) Image processing device and method, video display device, and recorded information reproduction device
JP4438795B2 (ja) 映像変換装置、映像表示装置、映像変換方法
JP4929819B2 (ja) 映像信号の変換装置及び方法
JP2007312317A (ja) 映像信号処理装置及び画像表示装置
US8836859B2 (en) Image processing apparatus, image processing method and image display apparatus
JP2009159332A (ja) 補間フレーム生成装置、補間フレーム生成方法及び放送受信装置
US20080129866A1 (en) Caption detection device, caption detection method, and pull-down signal detection apparatus
JP4001110B2 (ja) 走査変換装置
JPH02131689A (ja) フィールド内補間回路
US7079190B2 (en) Technique for determining the slope of a field pixel
JP2005045787A (ja) プログレッシブ及びインターレースビデオ信号の両方を生成するビデオ信号処理装置
US6417887B1 (en) Image display processing apparatus and method for converting an image signal from an interlaced system to a progressive system
US20050206786A1 (en) Motion compensation deinterlacer protection
JP3849299B2 (ja) テレビジョン受像機
US20110298977A1 (en) Video processing device
JP3692963B2 (ja) 映像源判別方法及び装置及び映像信号処理装置
US8134641B2 (en) Method and apparatus for processing video signal
US8872980B2 (en) System and method for accumulative stillness analysis of video signals
JP2009159321A (ja) 補間処理装置、補間処理方法及び映像表示装置
US8004606B2 (en) Original scan line detection

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A2

Designated state(s): JP

AL Designated countries for regional patents

Kind code of ref document: A2

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PT SE SI SK TR

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 2003585418

Country of ref document: JP

WWE Wipo information: entry into national phase

Ref document number: 2003746672

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 2003746672

Country of ref document: EP