Zero Shot Composed Image Retrieval Zs Cir On 1
평가 지표
R@5
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | R@5 |
|---|---|
| ldre-llm-based-divergent-reasoning-and | 55.57 |
| vision-by-language-for-training-free | 64.29 |
| covr-learning-composed-video-retrieval-from | 73.61 |
| semantic-editing-increment-benefits-zero-shot | 69.42 |
| reason-before-retrieve-one-stage-reflective | 57.68 |
| training-free-zs-cir-via-weighted-modality | 60.41 |
| isearle-improving-textual-inversion-for-zero | 54.05 |
| isearle-improving-textual-inversion-for-zero | 55.18 |
| compodiff-versatile-composed-image-retrieval | 57.61 |
| semantic-editing-increment-benefits-zero-shot | 57.42 |
| training-free-zs-cir-via-weighted-modality | 57.69 |
| reason-before-retrieve-one-stage-reflective | 67.25 |
| pretrain-like-you-inference-masked-tuning | 54.58 |
| magiclens-self-supervised-image-retrieval | 58.0 |
| ldre-llm-based-divergent-reasoning-and | 66.39 |
| magiclens-self-supervised-image-retrieval | 64.0 |
| language-only-efficient-training-of-zero-shot | 64.72 |
| imagescope-unifying-language-guided-image-1 | 67.54 |
| ldre-llm-based-divergent-reasoning-and | 55.13 |
| vision-by-language-for-training-free | 52.51 |
| compodiff-versatile-composed-image-retrieval | 54.36 |
| collm-a-large-language-model-for-composed | - |
| training-free-zs-cir-via-weighted-modality | 59.76 |
| magiclens-self-supervised-image-retrieval | 67.0 |
| reducing-task-discrepancy-of-text-encoders | 67.47 |
| imagine-and-seek-improving-composed-image | 58.82 |
| improving-composed-image-retrieval-via | 65.42 |
| pic2word-mapping-pictures-to-words-for-zero | 51.70 |
| collm-a-large-language-model-for-composed | - |
| pretrain-like-you-inference-masked-tuning | 58.87 |
| zero-shot-composed-image-retrieval-with | 53.42 |
| vision-by-language-for-training-free | 52.31 |
| isearle-improving-textual-inversion-for-zero | 55.69 |
| zero-shot-composed-text-image-retrieval | 68.88 |
| collm-a-large-language-model-for-composed | - |
| language-only-efficient-training-of-zero-shot | 53.25 |
| this-is-my-unicorn-fluffy-personalizing | 43.49 |
| magiclens-self-supervised-image-retrieval | 61.7 |
| imagine-and-seek-improving-composed-image | 70.07 |
| reducing-task-discrepancy-of-text-encoders | 56.17 |
| training-free-zs-cir-via-weighted-modality | 60.87 |
| zero-shot-composed-image-retrieval-with | 52.48 |
| context-i2w-mapping-images-to-context | 55.1 |
| reason-before-retrieve-one-stage-reflective | 54.54 |
| isearle-improving-textual-inversion-for-zero | 54.00 |
| semantic-editing-increment-benefits-zero-shot | 57.16 |