AI Paper Atlas

Research map / Video / Video generation

Chronologically Accurate Retrieval for Temporal Grounding of Motion-Language Models

Kent Fujiwara, Mikihiro Tanaka, Qing Yu

ECCV 2024 · Video › Video generation · approach: Text-to-motion diffusion · 8 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 shuffled-event texts as hard negatives in contrastive trainingTMR (DistilBERT) · HumanML3D motion-to-text retrieval, orig -> eventCAR accuracy: 64.81 -> 99.33Improved
add shuffled-event hard negativesTMR (DistilBERT) · HumanML3D, event -> eventR@1: 6.93 -> 8.90Improved
remove the VAE text encoder when fine-tuning with negatives (ablation)Ours t5-large · HumanML3D text-to-motion retrievalR@1: 6.57 -> 8.03Improved
keep VAE but remove motion reconstruction decoder (ablation)Ours t5-large · HumanML3D text-to-motion retrievalR@1: 6.57 -> 4.65Hurt
fine-tune CLIP text encoder via TMR without negatives (ablation)T2M-GPT · HumanML3D generationR-Precision Top-1: 0.489 -> 0.498Improved
add chronological negatives on top of tuned CLIP (ablation)T2M-GPT · HumanML3D generationFID: 0.107 -> 0.070Improved
add chronological negatives on top of tuned CLIP (ablation)ReMoDiffuse · HumanML3D generationR-Precision Top-1: 0.525 -> 0.525No clear effect
train text encoder with chronological negativesT2M-GPT · HumanML3D likelihood of true vs shuffled textfraction preferring true text: 61.9% -> 89.9%Improved
larger language model when fine-tuning (ablation)Motion-language retrieval · HumanML3D text-to-motion retrievalR@1: t5-large 8.03 vs CLIP 5.57 best-configImproved

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