sanowar-raihan / nerf-meta Goto Github PK

HI. Thanks for your pytorch code. Can you share the pre-trained meta models? Besides, I found the training is not very stable, sometimes the val PSNR keeps the same number while training and the test results are just some lumps.

I would like to ask about your experimental part in Single image view synthesis for ShapeNet, you are using only one image in the meta-learning stage under the setting of SV Meta, which means that only one image is used for the whole training process to generate the class 360 degree reconstruction effect for all instances？

Thank you for sharing this code in Pytorch version.
This would save many others including me.

I have one question about shapenet car dataset.
I tried to train this repository using shapenet/cars dataset and realized that some of the folders do not include images or transforms.json

I also leave the same question in the jax repository, but I wonder how you handle this problem.
link to the question: tancik/learnit#13 (comment)

Currently, I removed the corresponding folders from car_splits.json and rerun the code.

python shapenet_train.py --config ./configs/shapenet/cars.json

showing

Epoch: 1, val psnr: 20.808
Epoch: 2, val psnr: 21.266

Thanks in advance.

Hello, thank you for sharing the code for this wonderfull research! Is it possible to share the data for the Photorealism data used during the experiments?

Hi, I'm a beginnger for meta-learning. As I know, it is used for fitting model across different data. In each epoch for meta-learning, there are a batch of tasks and we should make copies of model for each task. For example, to meta-learn the shapenet dataset, in each epoch, we select several cars to process, thus the dimension of dataflow should be [b, ...], while b is the batch size. So, we need to make b copies of basic model so that each model can fit a particular car. However, I find that you only make one copy for the model in the inner loop(copy.deepcopy). Could you please give an explanation here? Do I misunderstand it?

Hello, thank you for releasing this implementation. I tried to use the scripts for view synthesis from a single image. I am running with CUDA 10.1 in a Ubuntu 16.04. I am using the lamps set of images and the lamps config script. However, when I end training, the resulting video is blank, with only a white void.

Excuse, I would like to ask about the construction process of this dataset. If I want to construct other classes objects of shapenet, what should I do? Thanks!