Digital Video Signal Modification TRIZ Cases

Spectral decomposition and noise reduction refine block prediction from reconstructed neighbors, improving coding efficiency with less extra signaling.

Dividing large coding units into sub-blocks improves intra prediction accuracy, coding efficiency, and reference pixel buffer use.

Filtered current and reference templates improve video block prediction, boosting compression efficiency without degrading reconstruction quality.

Varying CCSO parameters by filtering unit improves reconstructed video filtering quality while limiting bitstream signaling overhead.

Using neighboring samples and gradients, this case refines subblock motion vectors to improve video reconstruction without added memory cost.

Bitstream signaling of subpicture size and conformance windows lets decoders infer CTB splits at non-multiple boundaries after rearrangement.

Separate coding of significant and consecutive low-value MCA coefficients improves compression efficiency while preserving mura correction accuracy.

Look-ahead features and a neural model predict QP for intra and scene change frames, improving frame size control and video quality.

On-demand transcoding replaces stored bitrate ladders, cutting video storage and bandwidth while matching each client’s codec and network conditions.

Extended track selection and track group boxes move CMAF switching set signaling into ISOBMFF and support hierarchical track grouping.

Padding out-of-boundary reference blocks with motion-derived samples improves temporal prediction quality and video coding efficiency.

Combining non-linear and linear chroma prediction modes improves intra coding efficiency and lowers bit rate without hurting video quality.

Neighbor-based affine mode signaling improves video compression for zoom and rotation while limiting encoding and decoding complexity.

Constraining intra BC reference blocks to stay within inner or outer search areas avoids split on-chip/off-chip memory access while preserving compression quality.

Context from prior flags, block sizes, and neighboring blocks cuts signaling overhead in recursive intra region video coding.

Multiple direct chroma modes are derived from luma prediction and de-duplicated against default modes to improve coding efficiency.

Separable convolutions approximate multi-dimensional NN video filters, cutting compute load and memory bandwidth in coding.

Averaging two merge-mode motion vectors and pairing correction data expands prediction candidates while improving coding efficiency and decoding speed.

History counters are localized to CTU rows and synchronized between rows, enabling stable wavefront video decoding with parallel processing.

Resetting the HMVP buffer at tile row starts removes irrelevant motion vectors and improves video decoding compression efficiency.

Energy data carried through the streaming chain lets end devices request lower-consumption content profiles without limiting control to codecs alone.

Multiple filter shapes are selected per video block to improve filtering performance and coding efficiency without relying on one fixed shape.

A slice-header state flag signals picture header presence, cutting bit overhead while supporting multiple slices and clear picture boundaries.

Shared chroma residuals and adaptive Cb/Cr weights cut coding data while preserving image quality in video encoding and decoding.

Sampling-map motion vector prediction improves coding-block predictor accuracy while limiting extra data and coding complexity.

Boundary strength derived from adjacent CPR prediction modes improves deblocking at block edges while preserving compression efficiency and image quality.

Tuple-based DST-VII core generation cuts multiplication operations while preserving matrix-equivalent integer decoding results for video blocks.

Integer clipping and conditioned weights prevent accumulator overflow and preserve bitexact neural network coding across devices.

Selecting fG or fC interpolation filters from ISP status improves intra prediction efficiency while limiting block boundary artifacts.

Conditional slice header signaling omits redundant tile syntax elements, preserving decoder reconstruction while reducing bitstream overhead.

Separate non-VCL tracks for PH, APS, DCI, and OPI signaling reduce redundant VVC transmission and simplify decoder configuration.

Block-selective ICT improves RGB image and video coding by exploiting inter-component redundancy with lower complexity and low signaling overhead.

Selecting coding schemes from preceding image zone values cuts signaling cost and complexity while improving video compression.

CIIP-based block prediction improves video compression efficiency by reducing unnecessary signaling while balancing image quality and processing time.

Pixel-intensity-based deblocking adjusts β and tc in HDR video coding to reduce visible blocking artifacts without uniform filtering.

Tile-based texture and depth tracks let one stream support 3DoF and 3DoF+ rendering while cutting volumetric video bandwidth and transmission time.

Spatial-gradient PROF refines affine block prediction to improve compression efficiency while limiting motion estimation complexity in video coding.

A metadata box and configuration links let neural network bitstreams be stored and streamed in media files with better interoperability.

Short-term reference pictures and differential motion vectors improve bi-prediction coding efficiency while lowering signaling complexity.

Subblock-wise intra-prediction improves image coding accuracy by selecting adaptive block partitions and coding order with manageable complexity.

Selective template-matching reordering improves GPM coding efficiency for inter-predicted blocks while avoiding intra-prediction overhead.

Preserving AUD NAL units and filtering SEI messages by payload type keeps extracted VVC sub-bitstreams conformant.

Averaging merge-mode motion vectors and storing linked correction data improves prediction candidate selection with less encoding overhead.

Selective DMVD refines motion vectors only where needed, improving video coding efficiency and decoded quality without full decoding overhead.

Intra-block ARGB prediction combines DPCM, MED, and residual encoding to cut DDR image data traffic and power without losing pixels.

Geometry partition and HoG classifiers improve adaptive loop filter selection for reconstructed video blocks, reducing decoded distortion.

Template candidate lists let the decoder adapt intra prediction to quantization and block size, improving compression with limited processing.

Directional pre-processing preserves CTU boundary reference information, improving parallel video coding efficiency without losing intra prediction accuracy.

Neighbor-based merge candidate selection cuts prediction signaling and motion-information complexity while preserving inter prediction accuracy.

Using bypass coding for later CABAC bins cuts decoding complexity while preserving compression efficiency for syntax elements such as MVP and GBI.

Geometry-transformed reference neighborhoods expand template matching for intra-prediction, improving video compression with manageable complexity.

Filtered reference samples and inherited block parameters improve inter and intra prediction efficiency in video coding.

Sub-block partitioning and adaptive reduced-dimension non-separable transforms improve high-resolution image compression while limiting decoding complexity.

Adaptive reference regions and filtering modes improve intra prediction accuracy while limiting video encoding and decoding complexity.

Selective motion-candidate insertion and template-based refinement improve video coding quality while keeping prediction-list complexity manageable.

Similarity-threshold pruning removes redundant motion vector candidates before refinement, cutting video coding complexity while preserving prediction accuracy.

Simplified CIIP removes BDOF, shifts bi-prediction to uni-prediction, and unifies MPM handling to improve coding efficiency and decoding throughput.

A media-level VDII grouping structure avoids parsing every video track, enabling faster identification of tracks that share decoder initialization data.

Combined SAO flags signal common luma and chroma parameters to cut coding rate while reducing visual artifacts in reconstructed video.

Implicitly linking TU size to PU size cuts HEVC encoder complexity and signaling overhead while preserving coding efficiency.

Expanded spatial and secondary merge candidate lists improve inter-prediction and motion compensation for high-resolution video decoding.

Tailored transforms for L-shaped image partitions improve video compression efficiency while reducing encoding and decoding complexity.

Weighted multi-reference prediction improves high-resolution image coding accuracy while cutting signaling overhead in encoding and decoding.

A two-pass motion reconstruction flow prefetches reference blocks before decoder-side refinement, cutting memory latency for high-throughput video decoding.

Selective node splitting and motion compensation improve point cloud compression while limiting bandwidth, storage use, and decoding overhead.

Selective DMVR and BDOF flags at lower syntax levels cut decoding load while preserving picture quality and coding efficiency.

Quad-split intra subpartitions and a secondary transform improve coding efficiency, cut bit rate, and preserve local prediction modes.

Reverse last-significant-coefficient scanning cuts coding overhead and improves throughput for high bit depth, high bitrate video.

Block-based filter selection uses representative values from current and neighboring blocks to cut sample buffer needs while preserving image quality.

Non-differential coefficient coding and non-exponential clip index coding cut adaptive loop filter complexity and speed video encoding and decoding.

A two-layer point cloud codec separates lossy base data from lossless detail metadata to cut bit-rate while preserving reconstruction quality.

Adaptive MPM grouping and replacement of unavailable pixels improve intra prediction accuracy and video encoding efficiency.

A block map marks decoded blocks so image decoders reference only available neighbors, improving high-resolution compression and decode accuracy.

GCI syntax disables unneeded palette, intra, and transform features so video decoding keeps coding efficiency without extra processing.

Weight index derivation for bi-prediction improves video compression efficiency by refining affine merge motion candidates.

Dynamic non-zero coefficient limits let inverse LFNST decoding preserve coding quality while meeting worst-case multiplication constraints.

Decoded-block maps guide which neighboring regions are valid after block splitting, improving high-resolution image decoding efficiency and compression.

Boundary metadata added to free-view video files helps clients identify boundary cameras and render scene edges more accurately.

An edge server converts multi-view video segments on demand to match different display formats, cutting storage needs while preserving interoperability.

Motion vector offsets refine merge candidates to improve inter prediction accuracy and compression efficiency for high-resolution video decoding.

A reverse flag for last significant coefficient position cuts bitstream overhead and speeds high-bitrate video coding.

An encoder detects near-identical video frames and signals longer display duration to cut redundant encoding, bitrate, and encoding time.

Selective secondary transforms skip 4×4 and smaller video blocks to reduce coding complexity and memory load in high-resolution video processing.

Predefined scan-order partition assignment cuts tile and brick signaling bitrate while preserving flexible video coding layouts.

Non-adjacent video units expand IBC and IntraTMP candidate lists to improve coding efficiency without exhaustive template matching.

Obtuse-angle intra prediction improves reference sample use and coding efficiency for high-aspect-ratio image and video blocks.

Right-shift CCLM parameter derivation cuts table lookups, memory use, and hardware complexity in chroma prediction decoding.

Adaptive transform basis selection by block size cuts video coding load, processing time, circuit scale, and signaling cost.

Adaptive multi-tap CCSO uses neighboring luma samples to classify chroma offsets, reducing reconstruction error and improving coding efficiency.

Output layer set signaling lets layered video coding adapt picture resolution and reference resampling while cutting bandwidth and storage use.