AI Paper Atlas

Research map / Video / Video-language understanding

CoVR-2: Automatic Data Construction for Composed Video Retrieval

Lucas Ventura, Antoine Yang, Cordelia Schmid, Gül Varol

TPAMI 2024 · Video › Video-language understanding · approach: Text-video retrieval · 38 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 caption retrieval loss (ablation)CoVR-BLIP-2 · CIRR (trained on WebVid-CoVR)R@1: 34.43 -> 41.42Improved
add caption retrieval loss (ablation)CoVR-BLIP-2 · FashionIQR@10: 27.78 -> 36.81Improved
add caption retrieval loss (ablation)CoVR-BLIP-2 · CIRCOmAP@5: 25.06 -> 27.84Improved
add caption retrieval loss (ablation)CoVR-BLIP-2 · WebVid-CoVR-TestR@1: 60.09 -> 59.82No clear effect
use BLIP-2 cross-attention fusion instead of MLP fusion (ablation)CoVR-BLIP-2 · WebVid-CoVR-TestR@1: 51.88 -> 59.82Improved
finetune on WebVid-CoVR instead of frozen embeddings (ablation)CoVR-BLIP-2 · WebVid-CoVR-TestR@1: 45.66 -> 59.82Improved
combine visual and text instead of visual only (ablation)CoVR-BLIP-2 · WebVid-CoVR-TestR@1: 36.03 -> 59.82Improved
use finetuned MTG-LLM instead of rule-based modification text (ablation)CoVR-BLIP-2 · WebVid-CoVR-TestR@1: 39.08 -> 59.82Improved
use finetuned MTG-LLM instead of GPT-paraphrased rule text (ablation)CoVR-BLIP-2 · CIRRR@1: 32.89 -> 41.42Improved
iterate over target videos instead of triplets (ablation)CoVR-BLIP-2 · WebVid-CoVR-TestR@1: 53.79 -> 55.24Improved
add HN-NCE hard negative up-sampling (ablation)CoVR-BLIP-2 · WebVid-CoVR-TestR@1: 55.24 -> 59.82Improved
add CC-CoIR image data to WebVid-CoVR pretraining (ablation)CoVR-BLIP-2 · CIRR zero-shotR@1: 41.42 -> 43.74Improved

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