Image Retrieval On Fashion Iq

评估指标

(Recall@10+Recall@50)/2

Recall@10

评测结果

各个模型在此基准测试上的表现结果

			Paper Title	Repository
SPN4CIR (SPRC)	66.41	56.37	Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and Negatives
SPN4CIR	66.41	-	Improving Composed Image Retrieval via Contrastive Learning with Scaling Positives and Negatives
SPRC	64.85	54.92	Sentence-level Prompts Benefit Composed Image Retrieval
Candidate Set Re-ranking	62.15	51.17	Candidate Set Re-ranking for Composed Image Retrieval with Dual Multi-modal Encoder
RUTIR (BLIP B/16)	61.32	-	Ranking-aware Uncertainty for Text-guided Image Retrieval	-
CASE	59.73	48.79	Data Roaming and Quality Assessment for Composed Image Retrieval
CaLa	57.96	46.69	CaLa: Complementary Association Learning for Augmenting Composed Image Retrieval
BLIP4CIR+Bi	55.4	-	Bi-directional Training for Composed Image Retrieval via Text Prompt Learning
CLIP4Cir (v3)	55.36	-	Composed Image Retrieval using Contrastive Learning and Task-oriented CLIP-based Features
RUTIR (CLIP ResNet50)	55.27	-	Ranking-aware Uncertainty for Text-guided Image Retrieval	-
Css-Net	51.34	-	Collaborative Group: Composed Image Retrieval via Consensus Learning from Noisy Annotations	-
MUR (4*ResNet50)	50.61	-	Composed Image Retrieval with Text Feedback via Multi-grained Uncertainty Regularization
CLIP4Cir (v2)	50.03	-	Conditioned and Composed Image Retrieval Combining and Partially Fine-Tuning CLIP-Based Features	-
MUR	47.28	-	Composed Image Retrieval with Text Feedback via Multi-grained Uncertainty Regularization
CLIP4Cir	47.21	-	Effective Conditioned and Composed Image Retrieval Combining CLIP-Based Features	-
MMRet-MLLM	46.1	35.6	MegaPairs: Massive Data Synthesis For Universal Multimodal Retrieval
RTIC-GCN	40.64	-	RTIC: Residual Learning for Text and Image Composition using Graph Convolutional Network
CoSMo	39.45	-	CoSMo: Content-Style Modulation for Image Retrieval With Text Feedback	-
CurlingNet	38.45	-	CurlingNet: Compositional Learning between Images and Text for Fashion IQ Data
VAL w/ GloVe	35.38	-	Image Search With Text Feedback by Visiolinguistic Attention Learning	-

0 of 21 row(s) selected.