metadata
library_name: pytorch
RetinaFace is a state-of-the-art single-stage face detector that combines a robust backbone with multi-task learning to predict face locations, landmarks, and context-aware features simultaneously.
Original paper: RetinaFace: Single-Shot Multi-Level Face Localisation in the Wild
RetinaFace-Resnet50
This model uses RetinaFace with a ResNet-50 backbone, providing a strong trade-off between detection accuracy and computational efficiency. It is well suited for face detection and landmark localization in applications such as authentication, video surveillance, and human–computer interaction.
Model Configuration:
- Reference implementation: Official RetinaFace source code
- Original Weight: RetinaFace weight
- Resolution: 3x640x640
- Support Cooper version:
- Cooper SDK: [2.5.3]
- Cooper Foundry: [2.2]
| Model | Device | compression | Model Link |
|---|---|---|---|
| RetinaFace-Resnet50 | N1-655 | Amba_optimized | Model_Link |
| RetinaFace-Resnet50 | N1-655 | Activation_fp16 | Model_Link |
| RetinaFace-Resnet50 | CV7 | Amba_optimized | Model_Link |
| RetinaFace-Resnet50 | CV7 | Activation_fp16 | Model_Link |
| RetinaFace-Resnet50 | CV72 | Amba_optimized | Model_Link |
| RetinaFace-Resnet50 | CV72 | Activation_fp16 | Model_Link |
| RetinaFace-Resnet50 | CV75 | Amba_optimized | Model_Link |
| RetinaFace-Resnet50 | CV75 | Activation_fp16 | Model_Link |