PQ
A lossy compression technique that partitions high-dimensional vectors and quantizes each part separately, substantially reducing vector storage and distance computation costs; it is commonly combined with IVF.
Timeline (last 90 days)
ANN builds special index structures (such as HNSW's hierarchical navigable small world graph, IVF's inverted clustering, and PQ's product quantization compression) to sacrifice a small amount of recall accuracy in exchange for order-of-magnitude speedups.
Approximate nearest neighbor search (ANN) reduces retrieval complexity from O(N) for exact nearest neighbor to approximately O(log N) at the cost of a small amount of recall, making millisecond-level retrieval possible under registries of tens of thousands of entries.
HDR display technology uses the PQ (Perceptual Quantizer) transfer function instead of the traditional gamma curve; the PQ function is based on the absolute luminance perception threshold of the human eye and allocates more encoding precision to dark areas.
All Facts (3)
ANN builds special index structures (such as HNSW's hierarchical navigable small world graph, IVF's inverted clustering, and PQ's product quantization compression) to sacrifice a small amount of recall accuracy in exchange for order-of-magnitude speedups.
50%UnverifiedApproximate nearest neighbor search (ANN) reduces retrieval complexity from O(N) for exact nearest neighbor to approximately O(log N) at the cost of a small amount of recall, making millisecond-level retrieval possible under registries of tens of thousands of entries.
50%UnverifiedHDR display technology uses the PQ (Perceptual Quantizer) transfer function instead of the traditional gamma curve; the PQ function is based on the absolute luminance perception threshold of the human eye and allocates more encoding precision to dark areas.
50%