meta/omnivore 🔢🖼️📝 → ❓

▶️ 247 runs 📅 Jan 2022 ⚙️ Cog 0.1.3+shimmed 🔗 GitHub 📄 Paper ⚖️ License
image-classification multi-modal rgbd-image-classification video-classification visual-representation

About

A Single Model for Many Visual Modalities

Example Output

Output

[object Object]

Performance Metrics

3.90s Prediction Time
72.27s Total Time
All Input Parameters 
{
  "topk": "5",
  "image": "https://replicate.delivery/mgxm/adfc09d5-9c1c-4f1d-a567-a5da7abc7d10/library.jpg",
  "model_name": "omnivore_swinB"
}
Input Parameters
topk Type: integerDefault: 5Range: 1 - 10
Choose top k predictions to return.
image Type: string
input image
model_name Type: stringDefault: omnivore_swinB
Choose a model
Output Schema

Type: array • Items Type: object

Version Details
Version ID
79fa47c043835651e0d36fbebe3e3e1c24080818fa85ae1fa87f7537bfa7f37a
Version Created
January 29, 2022
Run on Replicate →