AI Paper Atlas

Research map / Video / Video segmentation and tracking

Compressive Quantization for Fast Object Instance Search in Videos

Tan Yu, Zhenzhen Wang, Junsong Yuan

ICCV 2017 · Video › Video segmentation and tracking · approach: Mask propagation for unsupervised video segmentation · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
jointly optimize quantization and hashing instead of k-means followed by ITQcompressive quantization · Groundhog Day (k=40, l=256)mAP: 0.523 -> 0.654Improved
jointly optimize quantization and hashing instead of k-means followed by ITQcompressive quantization · CNN2h (k=40, l=256)mAP: 0.578 -> 0.751Improved
binarize k-means codewords with ITQ (Baseline3) instead of keeping real-valued codewordsk-means codewords · CNN2h (k=100)mAP: 0.845 -> 0.687Hurt
use compressive quantization instead of real-valued k-means codewords (Baseline2)compressive quantization · CNN2h (k=100, l=512)mAP: 0.846 -> 0.847No clear effect
compress proposals into k=40 k-means codewords instead of exhaustive matchingBaseline1 · CNN2hmAP: 0.856 -> 0.817Mixed
increase codewords per video from 10 to 100 (ablation)Baseline2 · CNN2hmAP: 0.625 -> 0.845Improved
use binary codewords of 512 bits instead of real-valued 256-d codewordsBaseline2 · CNN2h (k=100)memory: 70.32 MB -> 4.39 MBImproved
use Hamming distance on binary codes instead of real-valued distanceBaseline2 · CNN2h (k=100)search time: 314 milliseconds -> 8 millisecondsImproved
use compressive quantization instead of exhaustive searchBaseline1 · CNN2hsearch time and memory: 10.48 s -> 0.008 s; 2.060 GB -> 4.39 MBImproved
use compressive quantization instead of k-means + CBE hashingk-means + CBE · CNN2h (k=100, l=512)mAP: 0.791 -> 0.847Improved
use compressive quantization instead of k-means + LSHk-means + LSH · Groundhog Day (k=100, l=128)mAP: 0.393 -> 0.635Improved
longer binary code from 128 to 512 bits (ablation)compressive quantization · CNN2h (k=100)mAP: 0.347 -> 0.847Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy