denglina

``` dummy_input = torch.rand((1, 3, 224, 224)) torch.onnx.export(model, dummy_input, save_onnx_path, export_params=True, verbose=True, do_constant_folding=True, opset_version=11) ``` export onnx error: Process finished with exit code 136 (interrupted by signal 8: SIGFPE)

Pool Mesh

2

Hi! Thanks for this awesome project. I trying to train dnsplatter，using custom data collected with an Orbber femto camera. Contain RGB and depth images. And then I used Spectacular AI...

Question about DTU dataset fuse_post.ply

2

```python python train.py -s ./DTU/scan24 -m output/date/scan24 --depth_ratio 1 -r 2 ``` ```python python render.py -r 2 --depth_ratio 1 --skip_test --skip_train -m output/date/scan24 --iteration 30000 ``` And the result of...

mid360+广角鱼眼相机建图上色精度

6

> 是的，一般希望LiDAR和相机的FoV重合得越多越好，这样其实标定也比较容易。Mid360更适合搭配广角鱼眼相机，而Avia则更适合搭配针孔相机。选型没问题的话，两种LiDAR都能得到不错的效果。 _Originally posted by @xuankuzcr in [#50](https://github.com/hku-mars/FAST-LIVO2/issues/50#issuecomment-2621556802)_ 郑博好，我尝试使用mid360+广角的鱼眼相机，运行fast-livo2的效果上色精度有较大偏差。 已经修改camera.yaml 中的cam_model: PolynomialCamera。 粉色是地面的颜色，墙面是白色的。从建图效果上来看墙面的一部分也上成粉色了。 是因为鱼眼自身的畸变原因吗？