Image Coding TRIZ Cases

Transmitting 3D patch borders instead of occupancy maps reduces artifacts in dynamic textured meshes while maintaining compression efficiency.

A 3D data encoding method generates bit sequences for N-ary tree nodes to indicate point presence in child regions.

Signaling error vectors between original and reconstructed patches fixes quantization shifts, improving mesh reconstruction accuracy.

Encoding system merges image and message data into a single integrated stream, eliminating multiple transmission channels while preserving original resolution.

Inverse wavelet transform on quantized level values reduces information volume while maintaining mesh reconstruction accuracy.

A 3D data decoding apparatus refines attribute and geometry frames using decoded target information to reduce coding distortion.

Segmenting image data into transform units allows parallel chrominance decoding, eliminating sequential buffering requirements.

Entropy coding compresses residual vectors to save storage space while maintaining distinct user copies for legal compliance.

Grouping prediction units into larger transform units adapts block sizes to image characteristics, improving compression ratios without excessive complexity.

A point cloud up-sampling method inserts neighboring points based on tangent plane proximity to fill under-sampled regions.

Embedded rendering unit converts RGB images to YUV color space and provides attributes directly to a video encoder.

A method preloads a target image thumbnail in memory and renders it on a display control within an editing interface.

Unified access unit merges geometry and attribute data to reduce processing complexity while enabling efficient multiplexing across multiple codecs.

Parallel slice processing eliminates serial bottlenecks in neural image compression, improving rate-distortion performance without increasing processing time.

A coding device acquires target point attribute data to generate position bit streams by applying dynamic prediction adjustments based on measured time intervals.

A compression value generator calculates dynamic values based on image area ratios to maintain constant visual fidelity across varying display sizes.

A mesh decoding device generates displacement via inverse wavelet transform and quantization.

Reshaping high-acuity regions before multiplexing avoids visible compression errors at region borders while reducing buffer sizes and decoding latency.

Dynamic color selection reduces image degradation while maintaining efficient data compression.

A point cloud generating section positions points at mesh surface intersections with specified resolution vectors.

Encoder packs 3D point cloud patches into image frames to reduce storage requirements while maintaining data integrity.

A point cloud encoding method uses neighboring attributes to predict current values for efficient data representation.

Universal encoder uses LSTM and binarizing layers to compress arbitrary image sizes, eliminating separate network training for different rates.

Segmenting JPEG decoding into DC and AC passes reduces processing time while eliminating browser flickering during web page loading.

A progressive subsampled transmission method sends low-resolution frames first to enable immediate client-side display and reconstruction.

Segmented coverage masks compress MSAA tile descriptor data, reducing memory traffic and system resource demand.