Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/OmniVL

OmniVL

Reported on 39 benchmarks across 8 tasks · 1 paper · 6 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision25 results

VideoonMSR-VTT
text-to-video R@1· uses extra data· 2022-09-15
47.8
best: 64 (GRAM)
SOTA
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
VideoonMSR-VTT
text-to-video R@10· uses extra data· 2022-09-15
83.8
best: 89.6 (VAST)
SOTA
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
VideoonMSR-VTT
text-to-video R@5· uses extra data· 2022-09-15
74.2
best: 84.3 (VAST)
SOTA
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video RetrievalonMSR-VTT
text-to-video R@1· uses extra data· 2022-09-15
47.8
best: 64 (GRAM)
SOTA
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video RetrievalonMSR-VTT
text-to-video R@10· uses extra data· 2022-09-15
83.8
best: 89.6 (VAST)
SOTA
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video RetrievalonMSR-VTT
text-to-video R@5· uses extra data· 2022-09-15
74.2
best: 84.3 (VAST)
SOTA
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
VideoonDiDeMo
text-to-video R@1· uses extra data· 2022-09-15
52.4
best: 74.2 (InternVideo2-6B)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
VideoonDiDeMo
text-to-video R@10· uses extra data· 2022-09-15
85.4
best: 94.2 (vid-TLDR (UMT-L))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
VideoonDiDeMo
text-to-video R@5· uses extra data· 2022-09-15
79.5
best: 91.2 (vid-TLDR (UMT-L))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
VideoonKinetics-400
Acc@1· 2022-09-15
79.1
best: 93.6 (OmniVec2)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
VideoonKinetics-400
Acc@5· 2022-09-15
94.5
best: 98.9 (TubeViT-H (ImageNet-1k))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video CaptioningonYouCook2
BLEU-3· 2022-09-15
12.87
best: 24.12 (UniVL + MELTR)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video CaptioningonYouCook2
BLEU-4· 2022-09-15
8.72
best: 18.2 (VAST)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video CaptioningonYouCook2
CIDEr· 2022-09-15
1.16
best: 116.4 (HowToCaption)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video CaptioningonYouCook2
METEOR· 2022-09-15
14.83
best: 22.56 (UniVL + MELTR)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video CaptioningonYouCook2
ROUGE-L· 2022-09-15
36.09
best: 47.04 (UniVL + MELTR)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video RetrievalonDiDeMo
text-to-video R@1· uses extra data· 2022-09-15
52.4
best: 74.2 (InternVideo2-6B)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video RetrievalonDiDeMo
text-to-video R@10· uses extra data· 2022-09-15
85.4
best: 94.2 (vid-TLDR (UMT-L))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Video RetrievalonDiDeMo
text-to-video R@5· uses extra data· 2022-09-15
79.5
best: 91.2 (vid-TLDR (UMT-L))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Zero-Shot Video RetrievalonMSR-VTT
text-to-video R@1· uses extra data· 2022-09-15
34.6
best: 55.9 (InternVideo2-6B)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Zero-Shot Video RetrievalonMSR-VTT
text-to-video R@10· uses extra data· 2022-09-15
66.6
best: 85.1 (InternVideo2-6B)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Zero-Shot Video RetrievalonMSR-VTT
text-to-video R@5· uses extra data· 2022-09-15
58.4
best: 78.3 (InternVideo2-6B)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Zero-Shot Video RetrievalonDiDeMo
text-to-video R@1· uses extra data· 2022-09-15
33.3
best: 57.9 (InternVideo2-6B)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Zero-Shot Video RetrievalonDiDeMo
text-to-video R@10· uses extra data· 2022-09-15
68.5
best: 85.1 (InternVideo2-1B)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Zero-Shot Video RetrievalonDiDeMo
text-to-video R@5· uses extra data· 2022-09-15
58.7
best: 80 (InternVideo2-6B)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526

Natural Language Processing10 results

Visual Question Answering (VQA)onMSRVTT-QA
Accuracy· uses extra data· 2022-09-15
0.441
best: 0.496 (VLAB)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Visual Question Answering (VQA)onMSVD-QA
Accuracy· uses extra data· 2022-09-15
0.51
best: 0.61 (VLAB)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Image Captioningonnocaps-val-out-domain
CIDEr· 2022-09-15
106.3
best: 124.8 (BLIP-2 ViT-G FlanT5 XL (zero-shot))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Image Captioningonnocaps-val-out-domain
SPICE· 2022-09-15
14.2
best: 15.1 (BLIP-2 ViT-G FlanT5 XL (zero-shot))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Image Captioningonnocaps-val-near-domain
CIDEr· 2022-09-15
108.3
best: 120.2 (BLIP-2 ViT-G FlanT5 XL (zero-shot))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Image Captioningonnocaps-val-near-domain
SPICE· 2022-09-15
14.9
best: 15.9 (BLIP-2 ViT-G FlanT5 XL (zero-shot))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Image Captioningonnocaps-val-overall
CIDEr· 2022-09-15
107.5
best: 121.6 (BLIP-2 ViT-G FlanT5 XL (zero-shot))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Image Captioningonnocaps-val-overall
SPICE· 2022-09-15
14.7
best: 15.8 (BLIP-2 ViT-G FlanT5 XL (zero-shot))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Image Captioningonnocaps-val-in-domain
CIDEr· 2022-09-15
104.6
best: 123.7 (BLIP-2 ViT-G FlanT5 XL (zero-shot))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Image Captioningonnocaps-val-in-domain
SPICE· 2022-09-15
15
best: 16.3 (BLIP-2 ViT-G FlanT5 XL (zero-shot))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526

Robots2 results

Activity RecognitiononSomething-Something V2
Top-1 Accuracy· 2022-09-15
62.5
best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Activity RecognitiononSomething-Something V2
Top-5 Accuracy· 2022-09-15
86.2
best: 96.3 (DejaVid)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526

Time Series2 results

Action RecognitiononSomething-Something V2
Top-1 Accuracy· 2022-09-15
62.5
best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526
Action RecognitiononSomething-Something V2
Top-5 Accuracy· 2022-09-15
86.2
best: 96.3 (DejaVid)
OmniVL:One Foundation Model for Image-Language and Video-Language Tasks arXiv:2209.07526