issues
search
duanyiqun
/
DiffusionDepth
PyTorch Implementation of introducing diffusion approach to 3D depth perception ECCV 2024
https://arxiv.org/abs/2303.05021
Apache License 2.0
306
stars
17
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Why is it that in the paper, rgb is input to get dense depth, while in testing code, rgb, dep, gt is input, which becomes depth completion
#50
air158
opened
3 months ago
0
Update README.md
#49
duanyiqun
closed
3 months ago
0
About DDIM loss on custom datasets
#48
Zray26
closed
9 months ago
9
model test
#47
Han-SHS
closed
11 months ago
6
Wrong result
#46
wangjiyuan9
closed
11 months ago
2
Question regarding implementation and memory usage in the DiffusionDepth model
#45
wangjiyuan9
closed
10 months ago
1
About shell command of kitti
#44
wangjiyuan9
closed
12 months ago
2
A question about the `CNNDDIMPipiline`
#43
zhanglk9
closed
1 year ago
0
About kitti eigen split
#42
wangjiyuan9
closed
12 months ago
4
module 'distutils' has no attribute 'version'
#41
drozzy
closed
10 months ago
1
Custom Testing
#40
holmes27
opened
1 year ago
0
For overfitting problem
#39
ZhengrongZhang
closed
10 months ago
3
Adding noise to refined_depth_t vs gt_depth_t
#38
unlugi
opened
1 year ago
1
Question about self-diffusion and depth encoder
#37
filbertsun
opened
1 year ago
0
Unstable Training without L2 loss
#36
unlugi
closed
1 year ago
4
Update README.md
#35
duanyiqun
closed
1 year ago
0
Missing install instructions - wrong python version and requirements.txt
#34
drozzy
closed
1 year ago
1
Conda env?
#33
drozzy
closed
10 months ago
0
Update README.md
#32
duanyiqun
closed
1 year ago
0
Question regarding overfit
#31
inkyusa
closed
10 months ago
1
Question about using ResNet18/50/101 backbone usage
#30
unlugi
closed
1 year ago
4
Update README.md
#29
duanyiqun
closed
1 year ago
0
About loss function
#28
HencyChen
closed
1 year ago
6
Mismatch of clamp value for decoder
#27
HencyChen
closed
1 year ago
2
Question about DDIM loss
#26
zyp-byte
opened
1 year ago
2
Comment out simple completion function to accelerate training speed.
#25
duanyiqun
closed
1 year ago
0
Long Training time for one epoch
#24
YBZh
closed
1 year ago
3
Update README.md Fix
#23
duanyiqun
closed
1 year ago
0
use MMDetection3D v1.0.0rc1 still have incompatible
#22
Huskie377
closed
1 year ago
0
attention of the MCDB module as mentioned in the paper
#21
Zjz999hbq
opened
1 year ago
3
Question about model train
#20
ipwjincheng
closed
10 months ago
2
mm version incompatiable error
#19
Zjz999hbq
opened
1 year ago
8
noise prediction
#18
xiangli93
closed
1 year ago
2
Depth AutoEncoder
#17
VitorGuizilini-TRI
closed
1 year ago
1
Question about ddim loss
#16
wyddmw
closed
1 year ago
2
A question about the pred noise.
#15
northagain
opened
1 year ago
1
Revert "Fix readme typo"
#14
duanyiqun
closed
1 year ago
0
Is there a plan to publish checkpoints?
#13
hutchinsonian
closed
1 year ago
2
Evaluation issue in HAHI head
#12
HencyChen
closed
1 year ago
2
A question about the DDIM loss.
#11
ganyz
closed
1 year ago
1
Training example command for NYU Depth v2
#10
ganyz
closed
1 year ago
1
Questions about dataset and results
#9
Sec996
opened
1 year ago
11
About depth mask in training
#8
HencyChen
closed
1 year ago
3
Is this code for depth estimation or completion?
#7
northagain
closed
1 year ago
0
Loss Difference between paper and code
#6
RickyYXY
closed
1 year ago
6
Do you provide a model checkpoint
#5
ridwan-salau
closed
1 year ago
1
Fix readme typo
#4
erjanmx
closed
1 year ago
1
Training encoder decoder
#3
dsshim0125
closed
1 year ago
2
cannot find the mmdet3d you provided, hope can update, thnanks~
#2
visonpon
closed
1 year ago
1
Question about diffusion
#1
XiangMochu
opened
1 year ago
3
Next