CJ Ye
In addition to well-aligned CAD models handcrafted by artists, we introduce two novel methods, DisCo and OccGOD. DisCo enables high-fidelity 3D shape reconstruction from limited sensor data. OccGOD detects full...
@oliver-batchelor @HLinChen I discovered that the error mentioned earlier was caused by the removal of some functions, namely rename and remove, from the [utils.py](https://github.com/baegwangbin/DSINE/blob/main/utils/utils.py). To address this issue, I have...
@imlixinyang Thanks for the great work! I am also wondering when will the downstream code and checkpoints will be released.
@zhouilu We've included a lightweight reimplementation of UniSDF in our [repo](https://github.com/hugoycj/Instant-angelo), feel free to give it a try. https://github.com/autonomousvision/sdfstudio/assets/40767265/607bc73a-fc66-4817-909e-40eb353d09ac
@Gofinge As part of my research, I have remapped Matterport3D's semantic labels to ScanNet 20 semantic labels with the addition of a ceiling label. This was done as a means...
That's great! I am excited to see how your experiments turn out and how these datasets contribute to the performance. By the way, I am adding more realistic and synthetic...
The preprocessed data have been added to the README, and the training log can be found at this [link](https://cuhko365-my.sharepoint.com/:u:/g/personal/118010378_link_cuhk_edu_cn/EeCdRrrBRh9IqcYwRBlR6fIBmA1aGQr1FoLn9xEPAcBS2Q?e=Hrqtfa). The evaluation results are summarized in the table below | Metric...
@DerKleineLi @mli0603 Some updates for Meeting Room training results at 60K steps. It's :   I use [BlenderNeuralangelo](https://github.com/mli0603/BlenderNeuralangelo) to manually regenerate the bounding_box and sphere. I have also modified...
Yeap. I follow the same steps in provided convert_tnt_to_json.py scripts, followed by calculating bounding box using Blender. One interesting thing I found for Meetingroom scene is the sparse point clouds...
I have also tried the colmap scripts using my own data. The results are stills far from good maybe due to the training time, but we could observe some fine...