AI Paper Atlas

Research map / Video / Action recognition

NSNet: Non-Saliency Suppression Sampler for Efficient Video Recognition

Boyang Xia, Wenhao Wu, Haoran Wang, Rui Su, Dongliang He, Haosen Yang, Xiaoran Fan, Wanli Ouyang

ECCV 2022 · Video › Action recognition · approach: Contrastive video representation learning · 21 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 the non-saliency suppression loss (ablation)Video Glimpse Module · ActivityNetmAP (%): 73.4 -> 73.8Improved
add non-saliency suppression pseudo labels (ablation)Frame Scrutinize Module · ActivityNetmAP (%): 68.4 -> 74.7Improved
add a non-salient class without NS supervision (ablation)Frame Scrutinize Module · ActivityNetmAP (%): 68.4 -> 68.6No clear effect
add a non-salient class without NS supervision (ablation)Video Glimpse Module · ActivityNetmAP (%): 73.4 -> 73.4No clear effect
use transformer encoder instead of 1D Conv in feature embedding (ablation)NSNet · ActivityNetmAP (%): 73.6 -> 75.5Improved
use transformer encoder instead of MLP in feature embedding (ablation)NSNet · ActivityNetmAP (%): 74.3 -> 75.5Improved
use NS objective instead of soft-label baseline+ for frame saliency (ablation)Frame Scrutinize Module · ActivityNetmAP (%): 73.7 -> 74.7Improved
use prototype-based instead of response-based guiding saliency score (ablation)Frame Scrutinize Module · ActivityNetmAP (%): 74.1 -> 74.7Improved
fuse FSM and VGM saliency instead of using FSM alone (10 frames) (ablation)NSNet · ActivityNetmAP (%): 74.7 -> 75.5Improved
sample 10 frames instead of 5 (ablation)NSNet · ActivityNetmAP (%): 74.9 -> 75.5Improved
use MobileNetv2 instead of ShuffleNetv2 as lightweight extractor (ablation)NSNet · ActivityNetmAP (%): 70.8 -> 75.5Improved
use EfficientNet-b0 instead of MobileNetv2 as lightweight extractor (ablation)NSNet · ActivityNetmAP (%): 75.5 -> 76.0Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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