paper-with-me

Papers

Contrast-Unity for Partially-Supervised Temporal Sentence Grounding

2025-02-18 · Haicheng Wang, Chen Ju, Weixiong Lin, Chaofan Ma, Shuai Xiao, Ya zhang, Yanfeng Wang

Temporal sentence grounding aims to detect event timestamps described by the natural language query from given untrimmed videos. The existing fully-supervised setting achieves great results but requires expensive annotation costs; while the weakly-supervised setting adopts cheap labels but performs poorly. To pursue high performance with less annotation costs, this paper introduces an intermediate partially-supervised setting, i.e., only short-clip is available during training. To make full use of partial labels, we specially design one contrast-unity framework, with the two-stage goal of implicit-explicit progressive grounding. In the implicit stage, we align event-query representations at fine granularity using comprehensive quadruple contrastive learning: event-query gather, event-background separation, intra-cluster compactness and inter-cluster separability. Then, high-quality representations bring acceptable grounding pseudo-labels. In the explicit stage, to explicitly optimize grounding objectives, we train one fully-supervised model using obtained pseudo-labels for grounding refinement and denoising. Extensive experiments and thoroughly ablations on Charades-STA and ActivityNet Captions demonstrate the significance of partial supervision, as well as our superior performance.

📄 PDF Abstract BibTeX arXiv:2502.12917

Code (0)

등록된 구현이 없습니다.

Tasks

Contrastive LearningDenoisingSentenceTemporal Sentence GroundingUnity

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Non-contrastive sentence representations via self-supervision

2023-10-26 · Marco Farina, Duccio Pappadopulo

Sample contrastive methods, typically referred to simply as contrastive are the foundation of most unsupervised methods to learn text and sentence embeddings. On the other hand, a different class of self-supervised loss …

SentenceSentence Embeddings

Constraint and Union for Partially-Supervised Temporal Sentence Grounding

2023-02-20 · Chen Ju, Haicheng Wang, Jinxiang Liu, Chaofan Ma 외

Temporal sentence grounding aims to detect the event timestamps described by the natural language query from given untrimmed videos. The existing fully-supervised setting achieves great performance but requires expensive…

SentenceTemporal Sentence Grounding

Unsupervised Representation Learning in Partially Observable Atari Games

2023-03-13 · Li Meng, Morten Goodwin, Anis Yazidi, Paal Engelstad

State representation learning aims to capture latent factors of an environment. Contrastive methods have performed better than generative models in previous state representation learning research. Although some researche…

Atari GamesRepresentation Learning

ESimCSE: Enhanced Sample Building Method for Contrastive Learning of Unsupervised Sentence Embedding

2021-09-09 · COLING 2022 10 · Xing Wu, Chaochen Gao, Liangjun Zang, Jizhong Han 외

Contrastive learning has been attracting much attention for learning unsupervised sentence embeddings. The current state-of-the-art unsupervised method is the unsupervised SimCSE (unsup-SimCSE). Unsup-SimCSE takes dropou…

Contrastive LearningData AugmentationSentenceSentence Embedding+4

Self-supervised Temporal Learning

2021-01-01 · Hao Shao, Yu Liu, Hongsheng Li

Self-supervised learning (SSL) has shown its powerful ability in discriminative representations for various visual, audio, and video applications. However, most recent works still focus on the different paradigms of spat…

Contrastive LearningRetrievalSelf-Supervised LearningVideo Classification+1