场景训练

[longzeyilang]

你好！，你在鱼眼场景进行训练嘛？我想要在图像尺寸128*128上进行训练，不知道效果怎么样

[meyiao]

@longzeyilang 我没有训练，模型直接拿来用就行，我测试下来感觉效果比orb, sift要好很多，比superpoint也要好一些。

[longzeyilang]

@meyiao 我将模型尺寸转化为128*128的
`import torch
from modules.xfeat import XFeatModel
import onnxruntime as ort
import onnx
import onnxsim

os.environ['CUDA_VISIBLE_DEVICES'] = '' #Force CPU, comment for GPU

set the model to evaluation mode

net = XFeatModel().eval()

load the pretrained weights

net.load_state_dict(torch.load("weights/xfeat.pt", map_location=torch.device('cpu')))

Random input

x = torch.randn(1, 1, 240, 320)

output_path = 'xfeat.onnx'

export to ONNX

torch.onnx.export(net, x, output_path, verbose=True,
input_names=['input'],
output_names=['output_feats', "output_keypoints", "output_heatmap"],
opset_version=11)
# Simplify model
onnx_model = onnx.load(output_path)
model_sim, check = onnxsim.simplify(onnx_model)
if check:
# output_path = output_path.replace(".onnx", "_sim.onnx")
onnx.save(model_sim, output_path)
print("Simplified ONNX model has been saved.")
print("ONNX model saved as xfeat.onnx")

check the onnx model with onnxruntime

ort_session = ort.InferenceSession("xfeat.onnx")
print("ONNX model loaded successfully")

outputs = ort_session.run(None, {"input": x.numpy()})

pytorch model outputs

torch_outputs = net(x)

compare the outputs

for i in range(len(outputs)):
print(f"onnx output shape {i}: {outputs[i].shape}")
print(f"torch output shape {i}: {torch_outputs[i].shape}")
print(f"Output {i} comparison: {torch.allclose(torch_outputs[i], torch.tensor(outputs[i]))}")
print(f'Output {i} max diff: {torch.max(torch.abs(torch_outputs[i] - torch.tensor(outputs[i])))}')
print("\n")
`
送进模型检测，实际配对点效果并不好，不知道原因在哪里

[meyiao]

@longzeyilang 我QQ很久没用了，你可以把图像对发出来，我有空的时候就试下

[longzeyilang]

@meyiao 我的qq邮箱： 1143059946@qq.com，你能发信给我嘛？我给你发些邮件