Coarse to Fine: Video Retrieval before Moment Localization
Not on the map. Why?
Abstract (arXiv)
The current state-of-the-art methods for video corpus moment retrieval (VCMR) often use similarity-based feature alignment approach for the sake of convenience and speed. However, late fusion methods like cosine similarity alignment are unable to make full use of the information from both query texts and videos. In this paper, we combine feature alignment with feature fusion to promote the performance on VCMR.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 9 of its 10 references that are in the map. First posted on arXiv 14 Oct 2021.
Builds on
9 of 10 references.
References · 9
- 2020HERO: Hierarchical Encoder for Video+Language Omni-representation Pre-trainingcited by 362
- 2020Span-based Localizing Network for Natural Language Video Localizationcited by 235
- 2021ConSERT: A Contrastive Framework for Self-Supervised Sentence Representation Transfercited by 231
- 2020TVR: A Large-Scale Dataset for Video-Subtitle Moment Retrievalcited by 215
- 2018QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehensioncited by 164
- 2021VALUE: A Multi-Task Benchmark for Video-and-Language Understanding Evaluationcited by 95
- 2021Video Corpus Moment Retrieval with Contrastive Learningcited by 60
- 2020A Hierarchical Multi-Modal Encoder for Moment Localization in Video Corpuscited by 26
- 2021Parallel Attention Network with Sequence Matching for Video Groundingcited by 22
Built on by
Citing papers per year: 2022: 1.
1 paper.