AI Paper Atlas

Research map / Detection and segmentation / CNN features

Patchwork: A Patch-Wise Attention Network for Efficient Object Detection and Segmentation in Video Streams

Yuning Chai

ICCV 2019 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 33 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
add Q-learned attention instead of random policy (ablation)Patchwork M=4 N=2 · ImageNet VID val (policy ablation, 1/4 cost)mAP@0.5: 52.5 -> 54.3Improved
add Q-learned attention instead of scanning policy (ablation)Patchwork M=2 N=1 vs M=4 N=2 · ImageNet VID val (policy ablation, 1/4 cost)mAP@0.5: 51.7 -> 54.3Improved
add Q-learned attention instead of random policy (ablation)Patchwork M=4 N=1 · ImageNet VID val (policy ablation, 1/16 cost)mAP@0.5: 37.0 -> 41.6Improved
add Q-learned attention instead of scanning policy (ablation)Patchwork M=4 N=1 · ImageNet VID val (policy ablation, 1/16 cost)mAP@0.5: 37.9 -> 41.6Improved
add Patchwork Cell to scanning policy (ablation)Patchwork · ImageNet VID val (Patchwork Cell ablation)mAP@0.5: 39.5 -> 51.7Improved
add Patchwork Cell to DQN policy (ablation)Patchwork · ImageNet VID val (Patchwork Cell ablation)mAP@0.5: 43.2 -> 54.3Improved
use Patchwork cells with random attention instead of single-frame modelPatchwork M=4 N=2 · ImageNet VID val (policy ablation, 1/4 cost)mAP@0.5: 49.3 -> 52.5Improved
use Patchwork M=4 N=2 at roughly 25% computation instead of full single-frame modelMobileNetV2 SSD · ImageNet VID valmAP@0.5: 54.7 -> 54.3No clear effect
reinvest saved compute in depth multiplier 1.4 and flip ensemblingPatchwork M=4 N=2 · ImageNet VID valmAP@0.5: 54.7 -> 58.7Improved
use Patchwork M=4 N=2 at about 1/4 computationMobileNetV2 segmenter · DAVIS 2016 unsupervisedJ mean: 63.6 -> 62.1Hurt
use Patchwork M=4 N=2 at about 1/4 computationMobileNetV2 segmenter · DAVIS 2016 unsupervisedF mean: 59.5 -> 58.8Hurt
use Patchwork M=4 N=1 instead of single-frame interval=16 delay=15 at similar costMobileNetV2 segmenter · DAVIS 2016 unsupervisedJ mean: 27.5 -> 42.2Improved

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