InstantMesh

InstantMesh copied to clipboard

编译中断

4

/home/zhangzhilong/tools/anaconda/envs/instantmesh/lib/python3.10/site-packages/torch/include/torch/csrc/profiler/util.h:34:10: 致命错误：x86intrin.h：No such file or directory 34 | #include | ^~~~~~~~~~~~~ 编译中断。 ninja: build stopped: subcommand failed.

huahai2022

推理速度疑问

1

请问在huggingface的demo上使用的哪个模型呢？我在A5000上使用instant-nerf-large进行推理，用时是huggingface上的两倍，而且生成质量不如huggingface的demo。

huahai2022

Issue about the rendering of filtered objaverse

10

Thanks for your great work! I wonder how you select the range of elevations and azimuths when rendering the filtered objaverse. This may influence the performance of the trained model....

wenqsun

Issue related to training

4

Hi @bluestyle97, I'm trying to train the model, but how do I set up the dataset, does the train.py download and preprocess the filtered objaverse dataset, or what do we...

kunalkathare

最低需要多少显存？24G能跑得了吗？

7

我用24G的显存，报错了，是不是显存不够？

ChenHuiLang

模型和代码文件欢迎发布到wisemodel.cn开源社区哦

LiuDQ-wm

纹理不够清晰

2

感谢大佬的工作，这是我目前见到的最好的模型了。比triposr， openlrm, crm都要好很多，方法也很直截了当，先用mvd生成多视图，然后再接上transformer去猜triplane tokens，然后再接上flexicube组成一套differentiable的pipeline，直接去优化mesh。不过我实际infer的时候，发现mesh虽然学的不错，但是并没有很好的保留原有的纹理特征。请问可以在我的数据集上finetune解决这个问题吗？可以给一些指导意见吗？ 感谢

LaFeuilleMorte

but are not expected and will be ignored. Please verify your config.json configuration file.

1

执行python run.py configs/instant-mesh-large.yaml examples/hatsune_miku.png --save_video命令后 Loading pipeline components...: 0%| | 0/8 [00:00

guoluzheng

执行python app.py后下载的文件放在哪里了，怎么删除啊，内存告急😅

boganLY

Question about parameters

4

I noticed that the texture quality of my output did not meet my expectations, especially when compared to the demo results which looked significantly better. Could you please help me...

masonintokyo