AI Paper Atlas

Research map / Detection and segmentation / CNN features

Sequence Level Semantics Aggregation for Video Object Detection

Haiping Wu, Yuntao Chen, Naiyan Wang, Zhaoxiang Zhang

ICCV 2019 · Detection and segmentation › CNN features · approach: Efficient DETR query design · 295 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 semantics aggregation within a single frame (ablation)single-frame Faster R-CNN · ImageNet VID valmAP: 73.62 -> 75.26Improved
add semantics aggregation within a single frame (ablation)single-frame Faster R-CNN · ImageNet VID val (fast motion)mAP: 51.53 -> 51.43No clear effect
extend semantic neighbor sampling to full video sequence (ablation)single-frame semantics aggregation · ImageNet VID valmAP: 75.26 -> 80.25Improved
extend semantic neighbor sampling to full video sequence (ablation)single-frame semantics aggregation · ImageNet VID val (fast motion)mAP: 51.43 -> 61.38Improved
use 21 frames instead of 5 for aggregation (ablation)SELSA · ImageNet VID valmAP: +1.04 mAPImproved
increase sampling stride from 1 to 10 (ablation)SELSA 21 frames · ImageNet VID valmAP: 77.02 -> 79.36Improved
shuffled sampling of semantic neighbors instead of stride 10 (ablation)SELSA 21 frames · ImageNet VID valmAP: 79.36 -> 80.25Improved
add data augmentation (ablation)SELSA ResNet-101 · ImageNet VID valmAP: +2.44 mAPImproved
add data augmentation (ablation)SELSA ResNeXt-101 · ImageNet VID valmAP: 83.11 -> 84.30Improved
add Seq-NMS post-processing (ablation)SELSA ResNet-101 · ImageNet VID valmAP: 82.69 -> 82.48Hurt
add Seq-NMS post-processing (ablation)SELSA ResNeXt-101 · ImageNet VID valmAP: 84.30 -> 83.73Hurt
use ResNeXt-101 instead of ResNet-101 backboneSELSA · ImageNet VID valmAP: 80.25 -> 83.11Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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