first commit

.gitattributes

@@ -33,3 +33,33 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+examples/demo16.jpg filter=lfs diff=lfs merge=lfs -text
+models/da3-base.axmodel filter=lfs diff=lfs merge=lfs -text
+models/da3-base.onnx filter=lfs diff=lfs merge=lfs -text
+examples/demo04.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo08.jpg filter=lfs diff=lfs merge=lfs -text
+models/da3mono-large.axmodel filter=lfs diff=lfs merge=lfs -text
+output-ax.png filter=lfs diff=lfs merge=lfs -text
+examples/demo19.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo02.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo10.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo11.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo15.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo12.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo20.jpg filter=lfs diff=lfs merge=lfs -text
+models/da3-small.axmodel filter=lfs diff=lfs merge=lfs -text
+output-onnx.png filter=lfs diff=lfs merge=lfs -text
+examples/demo05.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo06.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo14.jpg filter=lfs diff=lfs merge=lfs -text
+models/da3metric-large.onnx filter=lfs diff=lfs merge=lfs -text
+examples/demo13.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo17.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo09.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo18.jpg filter=lfs diff=lfs merge=lfs -text
+models/da3mono-large.onnx filter=lfs diff=lfs merge=lfs -text
+models/da3-small.onnx filter=lfs diff=lfs merge=lfs -text
+examples/demo01.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo03.jpg filter=lfs diff=lfs merge=lfs -text
+examples/demo07.jpg filter=lfs diff=lfs merge=lfs -text
+models/da3metric-large.axmodel filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,90 @@
----
-license: mit
----
+---
+license: bsd-3-clause
+language:
+- en
+base_model:
+- depth-anything/DA3-SMALL
+- depth-anything/DA3-BASE
+- depth-anything/DA3MONO-LARGE
+- depth-anything/DA3METRIC-LARGE
+pipeline_tag: depth-estimation
+tags:
+- Depth-Anything-3
+---
+
+# 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: 3.4
+
+## Convert tools links:
+
+For those who are interested in model conversion, you can try to export axmodel through 
+
+- [The repo of original](https://github.com/ByteDance-Seed/Depth-Anything-3)
+
+- [The repo of AXera Platform](https://github.com/AXERA-TECH/Depth-Anything-3.AXERA.git), which you can get the detial of guide
+
+- [Pulsar2 Link, How to Convert ONNX to axmodel](https://pulsar2-docs.readthedocs.io/en/latest/pulsar2/introduction.html) 
+
+
+## Support Platform
+
+- AX650
+  - [M4N-Dock(爱芯派Pro)](https://wiki.sipeed.com/hardware/zh/maixIV/m4ndock/m4ndock.html)
+  - [M.2 Accelerator card](https://axcl-docs.readthedocs.io/zh-cn/latest/doc_guide_hardware.html)
+- AX630C
+  - [爱芯派2](https://axera-pi-2-docs-cn.readthedocs.io/zh-cn/latest/index.html)
+  - [Module-LLM](https://docs.m5stack.com/zh_CN/module/Module-LLM)
+  - [LLM630 Compute Kit](https://docs.m5stack.com/zh_CN/core/LLM630%20Compute%20Kit)
+ 
+|Chips|Time|
+|--|--|
+|AX650|  ms |
+|AX630C|  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:
+
+![](examples/demo01.jpg)
+
+```
+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:
+
+![](output-ax.png)
+

models/da3-base.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02f0291bb629de91860a17174351c82825a975f4979100882399219f39764ea6
+size 139250301

models/da3-base.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a9193dca2e27df2fa5cd3bac074afcf3d662a9939869f4177aeacf4eba73061
+size 451237395

models/da3-small.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9563bff5bde0aff1ce69ae026691035207d4c813fc5190ab118adff8da77366
+size 40927267

models/da3-small.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82b0da30b20e48b968b718588a3a0cb296d0df5cc09327e0bd3301343edf9e9f
+size 124726109

models/da3metric-large.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f65d9248e36663e989c926653a6f411aecbd822504c4c1364be05f22174e0c9
+size 382403608

models/da3metric-large.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:500faa78858c1b646b0e39974128479bf313a0b10c6071b47161c34fc308c17f
+size 1334346787

models/da3mono-large.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d70bb9242c90ad9425733d8aacd27d18d520f2e5f6f15f86efd08d9035a8745
+size 382396604

models/da3mono-large.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:166946c445e7be229fa20a87ea445d9b2f11b484d171915a273a9757b60e1f01
+size 1334346787

python/infer.py

@@ -0,0 +1,50 @@
+import argparse
+import cv2
+import numpy as np
+import axengine as axe
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--img",
+        type=str,
+        required=True,
+        help="Path to input image.",
+    )
+    parser.add_argument(
+        "--model",
+        type=str,
+        required=True,
+        help="Path to axmodel model.",
+    )
+    
+    return parser.parse_args()
+
+
+def infer(img: str, model: str, viz: bool = False):
+    img_raw = cv2.imread(img)
+    image = cv2.cvtColor(img_raw, cv2.COLOR_BGR2RGB) 
+    orig_h, orig_w = image.shape[:2]
+    image = cv2.resize(image, (504,280))
+    image = image[None]
+
+    session = axe.InferenceSession(model)
+    
+    depth = session.run(None, {"img": image})[0]
+
+    depth = cv2.resize(depth[0, 0], (orig_w, orig_h))
+    depth = (depth - depth.min()) / (depth.max() - depth.min()) * 255.0
+    depth = depth.astype(np.uint8)
+
+    depth_color = cv2.applyColorMap(depth, cv2.COLORMAP_INFERNO)
+
+    combined_result = cv2.hconcat([img_raw,  depth_color])
+            
+    cv2.imwrite("output-ax.png", combined_result)
+    
+    return depth
+
+
+if __name__ == "__main__":
+    args = parse_args()
+    infer(**vars(args))

python/infer_onnx.py

@@ -0,0 +1,56 @@
+import argparse
+import cv2
+import numpy as np
+import onnxruntime as ort
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--img",
+        type=str,
+        required=True,
+        help="Path to input image.",
+    )
+    parser.add_argument(
+        "--model",
+        type=str,
+        required=True,
+        help="Path to ONNX model.",
+    )
+    
+    return parser.parse_args()
+
+
+def infer(img: str, model: str, viz: bool = False):
+    img_raw = cv2.imread(img)
+    image = cv2.cvtColor(img_raw, cv2.COLOR_BGR2RGB) 
+    orig_h, orig_w = image.shape[:2]
+    image = cv2.resize(image, (504,280))
+    mean = np.array([123.675, 116.28, 103.53],dtype=np.float32).reshape(1,1,3)
+    std = np.array([58.395, 57.12, 57.375],dtype=np.float32).reshape(1,1,3)
+
+    image = (image-mean)/std 
+    image = image.transpose(2,0,1)
+    image = image[None]
+
+    session = ort.InferenceSession(
+        model, providers=["CUDAExecutionProvider", "CPUExecutionProvider"]
+    )
+    depth = session.run(None, {"img": image})[0]
+
+    depth = cv2.resize(depth[0, 0], (orig_w, orig_h))
+    depth = (depth - depth.min()) / (depth.max() - depth.min()) * 255.0
+    depth = depth.astype(np.uint8)
+
+    depth_color = cv2.applyColorMap(depth, cv2.COLORMAP_INFERNO)
+
+    combined_result = cv2.hconcat([img_raw,  depth_color])
+            
+    cv2.imwrite("output-onnx.png", combined_result)
+    
+    return depth
+
+
+if __name__ == "__main__":
+    args = parse_args()
+    infer(**vars(args))

python/requirements.txt

@@ -0,0 +1,3 @@
+onnx
+onnxruntime
+opencv-python