mlsanityData to 7 Oct 2026

Exploiting Semantic Role Contextualized Video Features for Multi-Instance Text-Video Retrieval EPIC-KITCHENS-100 Multi-Instance Retrieval Challenge 2022

arXiv 2206.14381 ↗First posted 29 Jun 2022cs.CV cs.IR cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

In this report, we present our approach for EPIC-KITCHENS-100 Multi-Instance Retrieval Challenge 2022. We first parse sentences into semantic roles corresponding to verbs and nouns; then utilize self-attentions to exploit semantic role contextualized video features along with textual features via triplet losses in multiple embedding spaces. Our method overpasses the strong baseline in normalized Discounted Cumulative Gain (nDCG), which is more valuable for semantic similarity. Our submission is ranked 3rd for nDCG and ranked 4th for mAP.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 12 of its 15 references that are in the map. First posted on arXiv 29 Jun 2022.

Builds on

12 of 15 references.

References · 12

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Rescaling Egocentric Visioncited by 529
  3. 2019End-to-End Learning of Visual Representations from Uncurated Instructional Videoscited by 506
  4. 2020Multi-modal Transformer for Video Retrievalcited by 395
  5. 2019EPIC-Fusion: Audio-Visual Temporal Binding for Egocentric Action Recognitioncited by 226
  6. 2019Use What You Have: Video Retrieval Using Representations From Collaborative Expertscited by 214
  7. 2020Fine-grained Video-Text Retrieval with Hierarchical Graph Reasoningcited by 180
  8. 2019Fine-Grained Action Retrieval Through Multiple Parts-of-Speech Embeddingscited by 101
  9. 2018Dual Encoding for Zero-Example Video Retrievalcited by 85
  10. 2021On Semantic Similarity in Video Retrievalcited by 45
  11. 2022RoME: Role-aware Mixture-of-Expert Transformer for Text-to-Video Retrievalcited by 11
  12. 2022Semantic Role Aware Correlation Transformer for Text to Video Retrievalcited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 3; 2024: 0; 2025: 1; 2026: 1.

5 papers.

  1. 2025Seeing Culture: A Benchmark for Visual Reasoning and Groundingcited by 8
  2. 2023An Overview of Challenges in Egocentric Text-Video Retrievalcited by 1
  3. 2023UniUD Submission to the EPIC-Kitchens-100 Multi-Instance Retrieval Challenge 2023cited by 0
  4. 2023Towards Debiasing Frame Length Bias in Text-Video Retrieval via Causal Interventioncited by 0
  5. 2026Cultural Moment Benchmark: Evaluating Video Cultural Reasoning and Grounding in Southeast Asiacited by 0