Client devices compress video data based on real-time network bandwidth estimates to reduce central server congestion.
A video transcoding system adjusts i-frame frequency and frame size to match mobile playback capabilities.
A graphics-aware encoder adapts video compression settings based on detected content properties.
Selective block encoding reduces bit rate costs while maintaining video distortion resilience during primary slice loss.
Deblock control circuit selectively disables filtering based on content complexity to reduce processing load.
A parallel image decoding device uses predictive data amount calculation to determine block processing order.
A video encoding method selects optimal prediction directions by computing and comparing mode costs using bidirectional motion vectors.
A pseudo-analog transmission system scales video coefficients to retain spatial redundancy for graceful degradation in noisy channels.
Adaptive residual updating schemes partition video blocks to apply early sub-block transforms, reducing prediction errors and improving coding efficiency.
An adaptive pre-filter adjusts video complexity before encoding to optimize bit-rate allocation.
Modifying the quantization step allows rapid bit rate adjustment without retraining the neural network, saving time while maintaining encoding performance.
Motion detection adjusts temporal filtering strength across video frames, reducing processing power while preserving movement details and clarity.
An adaptive video filter uses edge detection signals to adjust deblocking strength across macroblocks.
Direct interpolation selects filter coefficients based on pixel phase to expand image resolution without fixed grid constraints.
A multi-core platform encodes digital video signals by processing groups of pictures in parallel across multiple processor units.
Reconstructs VLC table entries from a base array and difference indications to minimize storage overhead.
Prediction unit size mapping selects interpolation filters implicitly, eliminating explicit signaling overhead to reduce bit rate in HEVC.
Generates predictive coding parameters from enhancement layer pixels to create base layer prediction signals.
A transmitting apparatus separates buffering periods into encoding and transmission headers to maintain synchronous reproduction.
MxPEG video streams embed high resolution still images via difference frames, avoiding quality loss and bandwidth spikes.
Insert unique markers before NAL units in multimedia bitstreams to eliminate counters and extra memory needed for error concealment.
Video encoders adjust intra-frame distortion metrics via scale factors and offsets, resolving approximation errors that degrade measurement precision.
Segmenting picture headers from parameter sets reduces redundant bit rate usage while enhancing decoding reliability against transmission errors.
A scalable motion estimation method sets target complexity for macroblock partitions to adapt processing speed across varying hardware constraints.
Dynamic processing paths in the video transcoder adjust computational complexity to resolve picture quality versus processing time trade-offs.
Segmenting the decoder into independent error checking, concealment, and decoding units prevents single-point failures and boosts throughput.
Independent video and audio processors reattach timestamps based on frame periods to prevent lip sync deviations while adapting to packet size changes.
Merges short groups of pictures to enable parallel encoding threads, ensuring exact bit allocation prevents buffer violations during real-time video coding.
Partition-based filters reduce buffer capacity by applying renormalized partial coefficients near block boundaries.
Adaptive inter-layer filters reconstruct enhancement layers from base layer estimates, reducing bit overhead while maintaining high video quality.
One-dimensional filtering of reference samples removes outliers before intra prediction, improving accuracy while reducing computational load.
Precomputes collocated macroblock data from reference pictures, reducing computational complexity and memory overhead during B slice decoding.