Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Video-adverb retrieval with compositional adverb-action embeddings

About

Retrieving adverbs that describe an action in a video poses a crucial step towards fine-grained video understanding. We propose a framework for video-to-adverb retrieval (and vice versa) that aligns video embeddings with their matching compositional adverb-action text embedding in a joint embedding space. The compositional adverb-action text embedding is learned using a residual gating mechanism, along with a novel training objective consisting of triplet losses and a regression target. Our method achieves state-of-the-art performance on five recent benchmarks for video-adverb retrieval. Furthermore, we introduce dataset splits to benchmark video-adverb retrieval for unseen adverb-action compositions on subsets of the MSR-VTT Adverbs and ActivityNet Adverbs datasets. Our proposed framework outperforms all prior works for the generalisation task of retrieving adverbs from videos for unseen adverb-action compositions. Code and dataset splits are available at https://hummelth.github.io/ReGaDa/.

Thomas Hummel, Otniel-Bogdan Mercea, A. Sophia Koepke, Zeynep Akata• 2023

Related benchmarks

TaskDatasetResultRank
Adverb-to-video retrievalHowTo100M
mAP W56.7
7
Adverb-to-video retrievalAdverbs in Recipes
mAP W70.4
7
Adverb-to-video retrievalActivityNet
mAP W23.9
7
Adverb-to-video retrievalMSR-VTT
mAP (W)37.8
7
Adverb-to-video retrievalVATEX
mAP W29
7
Video-to-Adverb RetrievalHowTo100M
Acc-A81.7
7
Video-to-Adverb RetrievalAdverbs in Recipes
Acc-A87.4
7
Video-to-Adverb RetrievalActivityNet
Acc-A77.1
7
Video-to-Adverb RetrievalMSR-VTT
Acc-A (Adverb Retrieval)78.6
7
Video-to-Adverb RetrievalVATEX
Acc-A0.817
7
Showing 10 of 13 rows

Other info

Code

Follow for update