paper-with-me

Zero-Shot Transfer Image Classification 벤치마크

Zero-Shot Transfer Image Classification on ObjectNet

9개 결과 · ⬇ CSV · JSON

Accuracy (Private)

42.62 53.86 65.11 76.35 87.6 2021-02 2026-11 CLIP — 72.3 (2021-02-26) LiT-tuning — 81.1 (2021-11-15) CoCa — 82.7 (2022-05-04) LiT ViT-e — 84.9 (2022-09-14) PaLI — 42.62 (2022-09-14) LiT-22B — 87.6 (2023-02-10) EVA-CLIP-E/14+ — 79.6 (2023-03-27) InternVL-C — 80.6 (2023-12-21) EVA-CLIP-18B — 82.2 (2024-02-06) CLIP — 72.3 (2021-02-26) LiT-tuning — 81.1 (2021-11-15) CoCa — 82.7 (2022-05-04) LiT ViT-e — 84.9 (2022-09-14) LiT-22B — 87.6 (2023-02-10)
RankModel Accuracy (Private)Accuracy (Public)Top 5 Accuracy PaperCodeYear
1 LiT-22B 87.6–– Scaling Vision Transformers to 22 Billion Parameters lucidrains/flash-cosine-sim-attention 2023
2 LiT ViT-e 84.9–– PaLI: A Jointly-Scaled Multilingual Language-Image Model google-research/big_vision 2022
3 CoCa 82.7–– CoCa: Contrastive Captioners are Image-Text Foundation Models mlfoundations/open_clip · facebookresearch/multimodal · lucidrains/CoCa-pytorch · +3 2022
4 EVA-CLIP-18B 82.2–– EVA-CLIP-18B: Scaling CLIP to 18 Billion Parameters baaivision/EVA · baaivision/eva 2024
5 LiT-tuning 81.1 54.5– LiT: Zero-Shot Transfer with Locked-image text Tuning mlfoundations/open_clip · google-research/vision_transformer · google-research/big_vision · +2 2021
6 InternVL-C 80.6–– InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks opengvlab/internvl · opengvlab/internvl-mmdetseg 2023
7 EVA-CLIP-E/14+ 79.6–– EVA-CLIP: Improved Training Techniques for CLIP at Scale baaivision/eva · PaddlePaddle/PaddleMIX · Yui010206/CREMA · +1 2023
8 CLIP 72.3-– Learning Transferable Visual Models From Natural Language Supervision openai/CLIP · mlfoundations/open_clip · towhee-io/towhee · +79 2021
9 PaLI 42.62–58.35 PaLI: A Jointly-Scaled Multilingual Language-Image Model google-research/big_vision 2022
1–9 / 9 페이지당 10 20 50 100