AI Paper Atlas

Research map / Video / Video-language understanding

RGNet: A Unified Clip Retrieval and Grounding Network for Long Videos

Tanveer Hannan, Md Mohaiminul Islam, Thomas Seidl, Gedas Bertasius

ECCV 2024 · Video › Video-language understanding · approach: Long video understanding · 12 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
unify clip retrieval and grounding in one RG-Encoder instead of a disjoint pipeline (ablation)RGNet · Ego4D-NLQ valR1@0.3: 14.15 -> 18.28Improved
add learned sparsifier for frame selection (ablation)RGNet · Ego4D-NLQ valR1@0.3: 16.12 -> 17.80Improved
add retrieval token for clip features (ablation)RGNet · Ego4D-NLQ valR1@0.3: 17.80 -> 18.28Improved
add attention loss (ablation)RGNet · Ego4D-NLQ valR1@0.3: 16.21 -> 18.28Improved
add contrastive loss with long-video negative clip sampling (ablation)RGNet · Ego4D-NLQ valR1@0.3: 17.41 -> 18.28Improved
learned retrieval instead of heuristic dot-product retrieval (ablation)RGNet · Ego4D-NLQ valClip retrieval R1: 12.41 -> 25.01Improved
unified architecture improves both retrieval and oracle groundingRGNet · MADOracle grounding R1@0.3: 29.49 -> 33.42Improved
pre-train with NaQ annotations (ablation)RGNet · Ego4D-NLQ valR1@0.3: 18.28 -> 20.63Improved
lower frame rate as heuristic sparsification (ablation)RGNet · Ego4D-NLQ valgrounding accuracyHurt
increase clip length beyond 180s (MAD) or 48s (Ego4D) (ablation)RGNet · MAD / Ego4D-NLQgrounding accuracyHurt
increase decoder queries beyond 5 (ablation)RGNet · Ego4D-NLQ valLVTG performanceHurt
RGNet vs SOONet on MAD R5@0.3RGNet · MADR5@0.3: 19.64 -> 18.72Hurt

Later papers that compared with this method

5 later accepted papers report RGNet as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.

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