ViewFusion
ViewFusion copied to clipboard
Code of ViewFusion
For the 3d generation, I can get a good result if I use some simple object like a chair, but I got really bad result when I try to generate...
At least what model of graphics card is needed? How long is the inference time?
I tested the 3D reconstruction on the batman sample example provided and the render video and .ply look terrible. Any advice? https://github.com/Wi-sc/ViewFusion/assets/150967838/c1d30edb-dd8e-44b2-b6fc-0b146f0f5d63 https://github.com/Wi-sc/ViewFusion/assets/150967838/65cde044-6c71-4294-a5e2-87827006193b
Hello author, thanks for open source such an excellent job! I noticed that there doesn't seem to be any mention in the repository of how to use multi-view images as...
Hi, as per this constructor: class RealDataset(Dataset): def __init__(self, data_dir='/home/ubuntu/workspace/zero123/3drec/data/real_images/'): self.meta = [] for model_name in os.listdir(data_dir): meta_dict = { "model_name": model_name.split('.')[0], "inference_index": 0, "inference_image_path": f"{data_dir}/{model_name}", } self.meta.append(meta_dict) self.transform =...