https://github.com/BUAADreamer/SPN4CIR
공식 코드Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and Negatives
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
Given a query composed of a reference image and a relative caption, Composed Image Retrieval (CIR) aims to retrieve target images that are visually similar to the reference one but incorporate the changes specified in the relative caption. The bi-modality of the query provides users with more precise control over the characteristics of the desired image, as some features are more easily described with language, while others can be better expressed visually.
Zero-Shot Composed Image Retrieval (ZS-CIR) is a subtask of CIR that aims to design an approach that manages to combine the reference image and the relative caption without the need for supervised learning.
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and Negatives
Pretrain like Your Inference: Masked Tuning Improves Zero-Shot Composed Image Retrieval
Vision-by-Language for Training-Free Compositional Image Retrieval
MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions
Pic2Word: Mapping Pictures to Words for Zero-shot Composed Image Retrieval
CoVR-2: Automatic Data Construction for Composed Video Retrieval
CompoDiff: Versatile Composed Image Retrieval With Latent Diffusion
Language-only Efficient Training of Zero-shot Composed Image Retrieval
Composed Image Retrieval for Training-Free Domain Conversion
"This is my unicorn, Fluffy": Personalizing frozen vision-language representations
"This is my unicorn, Fluffy": Personalizing frozen vision-language representations
ImageScope: Unifying Language-Guided Image Retrieval via Large Multimodal Model Collective Reasoning
Knowledge-Enhanced Dual-stream Zero-shot Composed Image Retrieval
Training-free Zero-shot Composed Image Retrieval via Weighted Modality Fusion and Similarity
MegaPairs: Massive Data Synthesis For Universal Multimodal Retrieval
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.