AI Paper Atlas

Research map / Video / Video segmentation and tracking

Efficient Video Transformers with Spatial-Temporal Token Selection

Junke Wang, Xitong Yang, Hengduo Li, Li Liu, Zuxuan Wu, Yu-Gang Jiang

ECCV 2022 · Video › Video segmentation and tracking · approach: Self-supervised video representation learning · 83 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
use temporal rather than spatial selection at block 0 (same ~50% FLOPs cut) (ablation)STTS on MViT-B16 · Kinetics-400 (MViT-B16, ~50% GFLOPs reduction)top-1 accuracy: 76.2 -> 77.5Improved
add joint temporal (block 0) and spatial (block 4) selection instead of single-dimension (ablation)STTS on MViT-B16 · Kinetics-400 (MViT-B16, ~50% GFLOPs reduction)top-1 accuracy: 77.3 -> 77.5Improved
replace single-step with multi-step selection (0, 4, 8) (ablation)STTS on MViT-B16 · Kinetics-400 (MViT-B16, ~50% GFLOPs reduction)top-1 accuracy: 77.5 -> 76.6Hurt
use anchor-based instead of token-based spatial selection (ablation)STTS on MViT-B16-S0_0.5 · Kinetics-400 (MViT-B16, anchor-based spatial selection ablation)top-1 accuracy: 57.6 -> 76.2Improved
use anchors for random selection (ablation)Rand.-S0_0.5 · Kinetics-400 (MViT-B16, anchor-based spatial selection ablation)top-1 accuracy: 16.7 -> 75.6Improved
score on token-level rather than anchor-level features (ablation)STTS on MViT-B16-S0_0.5 · Kinetics-400 (MViT-B16, anchor-based spatial selection ablation)top-1 accuracy: 75.4 -> 76.2Improved
vary sigma 0.05 / 0.1 / 0.2 in perturbed Top-K (ablation)MViT-B16-T0_0.5 · Kinetics-400 (MViT-B16, sigma sweep)top-1 accuracy: 77.2, 77.3, 77.1No clear effect
learned differentiable Top-K selection instead of random selection (ablation)MViT-B16-T0_0.5 · Kinetics-400 (MViT-B16, sigma sweep)top-1 accuracy: 75.5 -> 77.3Improved
add STTS temporal selection at 0.6 ratio to VideoSwin-B (ablation)VideoSwin-B · Kinetics-400top-1 accuracy: 82.7 -> 81.4Hurt
add STTS temporal selection at 0.7 ratio to VideoSwin-B (ablation)VideoSwin-B · Something-Something-V2top-1 accuracy: 69.6 -> 68.7Hurt
add STTS (T0.9 S4.9) to MViT-B16 for less compute (ablation)MViT-B16 · Kinetics-400top-1 accuracy: 78.4 -> 78.1No clear effect
add STTS T0.6-S4.9 to MViT-B16 for speedMViT-B16 · Kinetics-400 (MViT-B16, inference speed)throughput (videos/s): 79.3 -> 129.7Improved

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