Digital Video Signal Modification TRIZ Cases

A multi-task neural network uses shared and QP-specific layers to reduce compression artifacts without separate models.

Two luma-based chroma models blend near a threshold, reducing artificial edges and improving color-channel correlation.

This video processing approach derives extended intra modes from basic modes to improve block prediction without signaling every mode.

This case uses layer-aware temporal sublayers to improve compression efficiency for high-resolution image and video data.

This case maps chrominance quantization offsets to transform unit sizes, improving coding efficiency while limiting bit stream growth.

This coding approach reuses spatial and historical motion information to improve inter prediction while reducing processing load.

A CNN-based filter uses quantization and block partition maps to improve reconstructed pictures without real-time training.

This video decoding case signals downsampling filters at key frames, then inherits them to reduce overhead across inter-coded GOP frames.

This case combines flexible VVC block partitioning, adaptive transforms, and quantization scaling to improve compression and visual quality.

This video decoding case maps wide-angle modes to non-wide modes, reusing coding tools to improve bit usage for non-square blocks.

Selective sample filtering improves video coding efficiency for high-resolution data.

Split constraints reduce luma-chroma dependency, latency, and buffer needs.

Local constant-value prediction reduces residuals at sharp image transitions.

This video decoding case scales blending by color component and sampling format to improve luma-chroma inter prediction efficiency.

Neighboring-block checks map usable modes directly and apply a fallback, reducing LUT memory overhead in intra prediction processing.

Offset key frames across quality streams enable frequent playback switching without increasing key frame frequency or stream size.

Spatial and temporal merge candidates are derived in parallel to simplify motion compensation logic and reduce memory access bandwidth.

Lossless phase detection data compression exploits channel correlation to reduce storage and transmission power in mobile imaging.

Multiple CfL prediction types map luma to neighboring chroma samples, reducing bandwidth and storage in high-resolution video.

IRAP and GDR flags coordinate random access and default pictures, supporting stable recovery when reference pictures are unavailable.

Out-of-range BVP coordinates are adjusted before cost-based reordering to improve intra block copy prediction efficiency.

This case applies CNN filters with quantization and block partition maps to improve reconstructed video while limiting decoding overhead.

This case merges candidate motion units across pixel samples to apply non-translational models with lower processing complexity.

Interpolated CPB parameters stabilize video coding across bit rates with less signaling overhead.

A multi-flag coefficient encoder uses modulo and remainder values to better exploit coefficient statistics and reduce video bandwidth.

Sample-wise filter selection uses block characteristics and partial reconstruction to improve video coding accuracy at lower bitrates.

Merging slice-specific controls and using override flags lets VVC decoding skip unused picture-header syntax.

This video coding case adapts filter strength and border count to content complexity, improving quality while reducing computation.

This case uses spatial IBC candidates, HVMP updates, and zero-vector padding to reduce memory and processing demands in video decoding.

This decoder case adapts filter characteristics to neighboring transform bases, reducing boundary errors and improving coding efficiency.

The case conditionally signals independent subpicture boundaries to improve bitstream utilization and video decoding efficiency.

This case uses asymmetric three-way block partitioning to preserve local adaptability while limiting recursive decoding complexity.

Successive parameter subsets and multi-tap luma prediction reconstruct chroma samples while reducing video transmission volume.

Selective bilateral filtering reduces intra-prediction artifacts in video coding.

This case uses progressive regional refresh across pictures to reduce I-picture bit requirements and improve encoding efficiency.

This decoding approach signals alternative chroma filters to improve compression and visual quality while managing ALF signaling complexity.

Each subblock uses collocated reference data and template matching to improve motion vectors while reducing video redundancy.

Shape-specific parameters improve intra prediction coding and signaling efficiency.

This encoder selects deblocking strength from adjacent block coding modes, balancing reconstructed-image quality with processing load.

This video coding case checks reference-block boundaries and excludes OOB samples to improve prediction with lower computational overhead.

Adjacent-block merge candidates streamline video inter prediction and motion compensation.

Reduced secondary transforms improve image compression while limiting data and computation.

This video coding case infers LFNST zero-out regions, limiting explicit coefficient-position signaling and reducing bitstream size.

The decoder adjusts deblocking, bilateral, SAO, or loop filtering using local depth and motion differences to improve video quality.

Ordered spatial, temporal, and non-neighboring candidates improve HD/UHD prediction efficiency.

Binary, tri, and quad splits adapt prediction to block size and shape, improving compression efficiency without excessive complexity.

Signaled phase offsets select filters for accurate luma and chroma upsampling in scalable video coding.

This case uses pixel and encoding features from prior pipeline stages to adjust bitrate in real time for consistent video quality.

Multiple resolution-based intra modes adapt to block complexity, reducing mode signaling bits while preserving prediction accuracy.

This case orders candidate motion vectors by motion information to reduce signaling bits while controlling list-construction complexity.