Open Vocabulary Object Detection On Lvis V1 0
评估指标
AP novel-LVIS base training
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | AP novel-LVIS base training |
|---|---|
| lami-detr-open-vocabulary-detection-with | 43.4 |
| region-aware-pretraining-for-open-vocabulary | 32.1 |
| object-aware-distillation-pyramid-for-open | 21.7 |
| x-paste-revisit-copy-paste-at-scale-with-clip | 21.4 |
| detecting-twenty-thousand-classes-using-image | 17.8 |
| ovmr-open-vocabulary-recognition-with-multi | 34.4 |
| prompt-pre-training-with-twenty-thousand-1 | 25.2 |
| simple-open-vocabulary-object-detection-with | 25.6 |
| regionclip-region-based-language-image | 17.1 |
| bridging-the-gap-between-object-and-image | 21.1 |
| zero-shot-detection-via-vision-and-language | 16.1 |
| codet-co-occurrence-guided-region-word-1 | 37.0 |
| clipself-vision-transformer-distills-itself | 34.9 |
| detection-oriented-image-text-pretraining-for | 40.4 |
| clim-contrastive-language-image-mosaic-for | 32.3 |
| zero-shot-detection-via-vision-and-language | 16.6 |
| ov-dquo-open-vocabulary-detr-with-denoising | 39.3 |
| zero-shot-detection-via-vision-and-language | 18.7 |
| open-vocabulary-object-detection-with | 22.4 |
| ov-dquo-open-vocabulary-detr-with-denoising | 29.7 |
| aligning-bag-of-regions-for-open-vocabulary | 22.6 |
| detect-every-thing-with-few-examples | 34.3 |
| regionclip-region-based-language-image | 22.0 |
| zero-shot-detection-via-vision-and-language | 26.3 |
| rtgen-generating-region-text-pairs-for-open | 30.2 |
| retrieval-augmented-open-vocabulary-object | 21.9 |
| contrastive-feature-masking-open-vocabulary | 33.9 |
| comprehensive-multi-modal-prototypes-are | 31.5 |