nianticlabs/acezero
[ECCV 2024 - Oral] ACE0 is a learning-based structure-from-motion approach that estimates camera parameters of sets of images by learning a multi-view consistent, implicit scene representation.
PythonNOASSERTION
Issues
- 0
error while loading model
#26 opened by xuyanging - 0
- 3
- 1
Questions about camera calibration
#24 opened by zhangwenniu - 2
Incomplete reconstruction
#20 opened by Chenswn - 2
The point cloud visualization shows that the reconstruction effect is too poor. Please tell me what I did wrong.
#23 opened by chenhui2016 - 2
- 3
- 2
RuntimeError: Re-rendering failed too often...
#10 opened by jinyummiao - 2
Install Error: Package not found
#22 opened by arqueomendez - 1
PLY Point Cloud File Generation
#21 opened by jiangyijin - 2
urllib.error.HTTPError
#19 opened by Chenswn - 3
ace_zero.py traceback
#12 opened by briancantwe - 0
- 1
RGBD as input?
#16 opened by BJHYZJ - 12
Errors running ace_zero.py with WSL2
#9 opened by gradeeterna - 2
Submodule Build error
#13 opened by 5p6 - 1
The speed does not match the paper.
#14 opened by ioir123ju - 1
- 1
ffmpeg path
#5 opened by thiesk - 1
ACE0 with Sparse View Inputs
#6 opened by hwanhuh - 3
Zoe-Depth module error?
#4 opened by shankar-anantak - 1
Multiple cameras , intrinsics?
#3 opened by antithing - 5
- 1
Novel View Synthesis Results of DROID-SLAM
#2 opened by TwiceMao