AI Paper Atlas

Research map / Video / Video-language understanding

G2L: Semantically Aligned and Uniform Video Grounding via Geodesic and Game Theory

Hongxiang Li, Meng Cao, Xuxin Cheng, Yaowei Li, Zhihong Zhu, Yuexian Zou

ICCV 2023 · Video › Video-language understanding · approach: Temporal video grounding · 43 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 geodesic-guided contrastive learning (GCL) (ablation)G2L · ActivityNet-CaptionsR@1 IoU=0.7: 28.62 -> 33.35Improved
add semantic alignment term in GCL (ablation)G2L · ActivityNet-CaptionsR@1 IoU=0.7: 30.73 -> 33.35Improved
add semantic uniformity term in GCL (ablation)G2L · ActivityNet-CaptionsR@1 IoU=0.7: 31.14 -> 33.35Improved
add semantic Shapley interaction (SSI) (ablation)G2L · ActivityNet-CaptionsR@1 IoU=0.7: 30.01 -> 33.35Improved
add GCL and SSI over the baseline (ablation)G2L · ActivityNet-CaptionsR@1 IoU=0.5: 45.74 -> 51.68Improved
use geodesic distance instead of Euclidean distance (ablation)G2L · ActivityNet-CaptionsR@1 IoU=0.5: 47.83 -> 51.68Improved
use geodesic distance instead of timestamp distance (ablation)G2L · ActivityNet-CaptionsR@1 IoU=0.5: 49.85 -> 51.68Improved
use geodesic distance instead of cosine distance (ablation)G2L · ActivityNet-CaptionsR@1 IoU=0.7: 30.09 -> 33.35Improved
add GCL and SSI training objectivesBase model · ActivityNet-Captionstraining iteration time: 0.39s -> 0.84sHurt
add GCL and SSI training objectivesBase model · ActivityNet-Captionsinference time: 40s -> 43sHurt
G2L versus best prior method on TACoSG2L · TACoSR@5 IoU=0.5: 50.00 -> 49.86Mixed

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