Depth-Anything-3

This version of Depth-Anything-3 has been converted to run on the Axera NPU using w8a16 quantization.

This model has been optimized with the following LoRA:

Compatible with Pulsar2 version: 5.0-patch1

Convert tools links:

For those who are interested in model conversion, you can try to export axmodel through

• The repo of original

• The repo of AXera Platform, which you can get the detial of guide

• Pulsar2 Link, How to Convert ONNX to axmodel

Support Platform

• AX650
  • M4N-Dock(爱芯派Pro)
  • M.2 Accelerator card
• AX637

Chips	Models	Time
AX650	da3-base	67.341 ms
AX650	da3-small	22.768 ms
AX650	da3metric-large	217.577 ms
AX650	da3mono-large	217.615 ms
AX637	da3-base	174.100 ms
AX637	da3-small	75.802 ms
AX637	da3metric-large	698.765 ms
AX637	da3mono-large	697.891 ms

How to use

Download all files from this repository to the device

python env requirement

pyaxengine

https://github.com/AXERA-TECH/pyaxengine

wget https://github.com/AXERA-TECH/pyaxengine/releases/download/0.1.3.rc2/axengine-0.1.3-py3-none-any.whl
pip install axengine-0.1.3-py3-none-any.whl

others

Maybe None.

Inference with AX650 Host, such as M4N-Dock(爱芯派Pro)

Input image:

root@ax650:~/AXERA-TECH/Depth-Anything-3# python3 python/infer.py --model models/da3metric-large.axmodel --img examples/demo01.jpg
[INFO] Available providers:  ['AxEngineExecutionProvider']
[INFO] Using provider: AxEngineExecutionProvider
[INFO] Chip type: ChipType.MC50
[INFO] VNPU type: VNPUType.DISABLED
[INFO] Engine version: 2.12.0s
[INFO] Model type: 2 (triple core)
[INFO] Compiler version: 3.3 ae03a08f
root@ax650:~/AXERA-TECH/Depth-Anything-3# ls

Output image: