9796l

when i test the image,the result of the depth is black

1

when i use custom dataset (my own dataset) to train, after training ,i used the weight to test,python test_simple.py,but i got the bad result,the mono depth image is black ,when...

hello,after training the model,i used test_simple.py to test the model,but i do not know how to test many frames at the time? i can only use one target frame and...

excuse me, i use other RGBD outdoor dataset,besides,the resolution is the same as NYU,i use the same training method as NYU in this code.But the loss function is not converage.becuse...

RuntimeError: Given groups=1, weight of size [256, 128, 1, 1], expected input[1, 98, 176, 144] to have 128 channels, but got 98 channels instead

1

excuse me,i used datasets which are same with nyu,but when i train the dataset,the problem is RuntimeError: Given groups=1, weight of size [256, 128, 1, 1], expected input[1, 98, 176,...

excuse me,in the train_test_input/*.txt,bathroom/rgb_00045.jpg bathroom/sync_depth_00045.png 518.8579,in the last column,i do not know the meaning of 518.8579

Hello,i am very grateful for your work. I want to know how to generate red triangulations which represent pose in reconstructed scene in your github code. I want to visualize...

run demo but work poorly

1

hello， i am very grateful for your work.but when i run demo,the result is poor as below ![Uploading 2024-06-12 11-23-55 的屏幕截图.png…]()

Support for Ubuntu 18.04

3

hello , i am very grateful your work. when i run demo using pixi, i guess the version of pixi and ubuntu18.04 is conflict. when ubuntu18.04 cuda 11.3,what is the...

dear writers, i am very grateful for your successful work,i would like to know about that if the scene is multi-person,does the SMPLer work well ? and can the SMPLer...

Reproducing the demo works poorly

1

Hello, i am very grateful for your work.I work the demo of build.md,but the result is poor.