LHXhh

Results 7 comments of LHXhh

@cdcseacave Hello, have you implemented it yet“is there a way to export the point-cloud and/or mesh scaled/transformed back to the original coordinates?”

![1](https://github.com/nerfstudio-project/nerfstudio/assets/102360620/5aaf10a3-7202-4568-88a0-667720c5345d)

Thank you very much! ------------------ 原始邮件 ------------------ 发件人: ***@***.***>; 发送时间: 2022年10月5日(星期三) 晚上6:39 收件人: ***@***.***>; 抄送: ***@***.***>; ***@***.***>; 主题: Re: [Xharlie/pointnerf] RuntimeError (Issue #57) The supplied script files (e.g. lego_test.sh) have a line:...

I have the same question. Have anyone succeeded?

Thank you very much! ------------------ 原始邮件 ------------------ 发件人: "yenchenlin/nerf-pytorch" ***@***.***>; 发送时间: 2022年9月9日(星期五) 中午11:13 ***@***.***>; ***@***.******@***.***>; 主题: Re: [yenchenlin/nerf-pytorch] Extracting geometry from a NeRF (Issue #68) I extracted the mesh based on this file. You...

很感谢您的回复,我想请问下得到的这个verts_tex是每个顶点的(1,num_points, 7)特征向量,我该怎么提取对应的顶点颜色呢,就是不太清楚verts_tex怎么用?真诚地希望能够得到您的帮助 `with torch.no_grad(): vertices, faces, v, psr_grid, points = sap_generate(dpsr, psr2mesh, inputs, center, scale) verts_tex = grid_interp(tex_grid, v) sample_verts = v.squeeze(0).detach().cpu().numpy() save_verts = vertices.squeeze(0).detach().cpu().numpy() np_faces = faces.squeeze(0).detach().cpu().long().numpy() save_mesh =...

好的,很感谢。但是当我将verts_tex的漫反射和镜面反射分别提取出来时,发现在训练过程的一半时还有颜色信息,虽然效果很糟糕,但是训练结束时颜色值归0了。模型几何还可以,但没有了顶点颜色。我是完全遵照源代码和所给的数据集进行训练的,我看论文上的texture mesh还可以,您知道是什么缘故呢?抱歉打扰您了 训练到一半 ![image](https://github.com/l1346792580123/diff/assets/102360620/0acf8f02-8827-43aa-bff5-8d79d8da5a77) 训练结束时 ![image](https://github.com/l1346792580123/diff/assets/102360620/e62ccf53-e525-4c21-9f8f-092df81a69c3)