paper
1.[Paper] SINC: Spatial Composition of 3D Human Motions for Simultaneous Action Generation
그래도 그동안 논문을 읽지 않은 것은 아닌데 ... velog에서의 논문 리뷰는 꽤 오랫동안 취었네 ^^ ... 앞으로 일주일에 한 논문 리뷰하기 프로젝트 시작!
2.Learning to Prompt for Vision-Language Models
Publish: IJCV 2021 YEAR: 2021 LINK: https://arxiv.org/abs/2109.01134 Summary CLIP: 이전 representation learning method 와는 다르게, vision, language를 하나의 feature space 상에 align 이를 통해 downstream task에서 ...
3.LoCoOp: Few-Shot Out-of-Distribution Detection via Prompt Learning
Publish: NeurIPS 2023 Year: 2023 LINK: https://arxiv.org/abs/2306.01293 Summary few-shot out-of-distribution detection을 위한 새로운 vision-language prompt learning approach 제안 few-shot OoDD: 일부의 ID i...
4.Test-Time Linear Out-of-Distribution Detection
Publish: CVPR 2024 Year: 2023 LINK: https://openaccess.thecvf.com/content/CVPR2024/papers/FanTest-TimeLinearOut-of-DistributionDetectionCVPR2024_paper.pdf Summary 최근 OOD detection algorithms에 의해 만들어진...
5.BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning
Publish: CVPR 2024 Year: 2024 LINK: https://openaccess.thecvf.com/content/CVPR2024/papers/LiangBadCLIPDual-EmbeddingGuidedBackdoorAttackonMultimodalContrastiveLearningCVPR2024paper.pdf Summary CLIP과 ...