AI Paper Atlas

Research map / Video / Video-language understanding

Visual Relation Grounding in Videos

Junbin Xiao, Xindi Shang, Xun Yang, Sheng Tang, Tat-Seng Chua

ECCV 2020 · Video › Video-language understanding · approach: Temporal video grounding · 46 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 message passing between subject and object (ablation)vRGV · ImageNet-VidVRDAcc_R (%): 17.63 -> 24.58Improved
add clip-level attention (hierarchical structure) (ablation)vRGV · ImageNet-VidVRDAcc_R (%): 20.58 -> 24.58Improved
add temporal attention unit (ablation)vRGV · ImageNet-VidVRDAcc_R (%): 7.06 -> 24.58Improved
vary temporal attention threshold from 0.01 to 0.05 (ablation)vRGV · ImageNet-VidVRDAcc_R (%): 24.51, 24.76, 24.43, 24.58, 22.94No clear effect
link regions with explicit object locations (post-linking) instead of greedy attention linking (ablation)vRGV · ImageNet-VidVRDAcc_R (%): 20.54 -> 24.58Improved
link regions with explicit object locations (post-linking) in co-occurrence baseline (ablation)Co-occur · ImageNet-VidVRDAcc_R (%): 12.80 -> 16.48Improved
use vRGV instead of T-Rank V1grounding · ImageNet-VidVRDAcc_R (%): 3.99 -> 24.58Improved
use vRGV instead of Co-occurgrounding · ImageNet-VidVRDAcc_R (%): 16.48 -> 24.58Improved
concatenate trajectory pairs (T-Rank V2 vs V1)T-Rank · ImageNet-VidVRDAcc_O (%): 10.68 -> 7.35Hurt
use vRGV instead of Co-occur in zero-shotgrounding · ImageNet-VidVRD zero-shotAcc_R (%): 7.38 -> 10.27Improved
move from static to dynamic relationsCo-occur · ImageNet-VidVRDAcc_R (%): 21.4 -> 13.81Hurt

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