Image Retrieval On Cirr
Metrics
(Recall@5+Recall_subset@1)/2
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | (Recall@5+Recall_subset@1)/2 |
|---|---|
| target-guided-composed-image-retrieval | 75.6 |
| sentence-level-prompts-benefit-composed-image | 82.66 |
| improving-composed-image-retrieval-via | 82.69 |
| megapairs-massive-data-synthesis-for | - |
| data-roaming-and-early-fusion-for-composed | 78.25 |
| sentence-level-prompts-benefit-composed-image | 81.39 |
| conditioned-and-composed-image-retrieval | 69.09 |
| cala-complementary-association-learning-for | 78.74 |
| data-roaming-and-early-fusion-for-composed | 77.5 |
| composed-image-retrieval-using-contrastive | 75.10 |
| effective-conditioned-and-composed-image | 63.87 |
| bi-directional-training-for-composed-image | 72.59 |
| candidate-set-re-ranking-for-composed-image | 80.9 |
| vista-visualized-text-embedding-for-universal | 75.9 |
| image-retrieval-on-real-life-images-with-pre | 45.88 |
| improving-composed-image-retrieval-via | 82.69 |
| artemis-attention-based-retrieval-with-text-1 | 43.05 |