issues
search
WangYueFt
/
detr3d
MIT License
744
stars
139
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
what is meaning of self.query_embedding = nn.Embedding(self.num_query, self.embed_dims * 2)
#72
hitbuyi
opened
2 months ago
0
about the result of mAOE
#71
kronaldo123
opened
7 months ago
0
Incompatible package versions
#70
JoshuaScheuplein
opened
8 months ago
0
Questions about comprehension on loss function equation
#69
ChunZheng2022
opened
10 months ago
0
Single-Camera Images
#68
becauseofAI
opened
11 months ago
0
Path problem
#67
RhythmOfTheRain-Byte
opened
1 year ago
0
Welcome update to OpenMMLab 2.0
#66
vansin
opened
1 year ago
0
Training the model does not converge 0mAP 0 NDS score
#65
ruben-poghosyan
opened
1 year ago
0
Use DETR3D to train my own dataset
#64
zhangbaoj
opened
1 year ago
1
Question about NMSFreeCoder
#63
george0407
opened
1 year ago
1
Load ckpts/pillar.pth but shows size mismatch for pts_voxel_encoder, cannot reproduce the baseline result
#62
chaoyiwang09
closed
1 year ago
1
support detr3d in mmdet3d-dev-1.x (v1.1)
#61
ZLTJohn
opened
1 year ago
0
detr3d for mmdet3d v1.0.0
#60
ZLTJohn
opened
1 year ago
0
How to visualize detection results on images like in the paper
#59
cgl-cell
opened
1 year ago
0
Car detection performance of checkpoint on nuscenes mini is poor
#58
hrz2000
closed
1 year ago
2
Question on nan
#57
xyupeng
opened
1 year ago
0
About the vov-99 pretrained checkpoint
#56
zw615
opened
1 year ago
0
Where is the python file named pillar_cosine.py ?
#55
quanh1990
opened
1 year ago
1
Did you tried tempotal fusion?
#54
applezy8866
opened
1 year ago
0
RuntimeError: shape '[1, 4, 1, 4, 4]' is invalid for input of size 16
#53
shb9793
opened
1 year ago
0
How to reproduce the result of ResNet50 version?
#52
increase24
opened
1 year ago
1
Raise ValueError: matrix contains invalid numeric entries
#51
nacayu
opened
1 year ago
1
how to get your pretrained fcos3d checkpoint
#50
BoomSky0416
opened
1 year ago
1
About the abnormal mASE and mASE
#49
zhanghan0112
opened
1 year ago
1
Differenct coordinate setting in gt_bboxes and bboxes_preds
#48
nacayu
opened
1 year ago
3
two stage results
#47
HaoZhang534
opened
1 year ago
0
Questions about grid mask.
#46
cxmomo
opened
1 year ago
1
Knowledge Distillation
#45
AnnieYu0401
opened
2 years ago
0
Where can I find a pre-training model for Resent50?
#44
LuRight
opened
2 years ago
0
'ConfigDict' object has no attribute 'device'
#43
jongwoopark7978
opened
2 years ago
0
Detr3D is not in the models registry
#42
guillemdelgado
opened
2 years ago
3
Training diverge for different setting
#41
bronyayang
opened
2 years ago
0
About DD3D models
#40
XuyangBai
closed
2 years ago
3
Question regarding CBGS sampling
#39
pangsu0613
opened
2 years ago
0
no test config works
#38
Mo-Farazi
opened
2 years ago
3
Perfomance on val and test set.
#37
XuyangBai
closed
2 years ago
2
how to reproduce the result ?
#36
babyporgrammer
opened
2 years ago
1
Modify the backbone network to resnet50, the model does not converge！
#35
Suodislie
opened
2 years ago
2
Why you make pred box z = z - h/2 in head's get_bboxes
#34
zhanglazhuan
opened
2 years ago
2
can you provide the ensemble code?
#33
TianhaoFu
opened
2 years ago
1
A question about GPU-util and training time(detr3D)
#32
Etah0409
closed
2 years ago
4
mAOE metric abnormal
#31
kaixinbear
closed
2 years ago
7
confused with the evalution results from val-set
#30
Yqmtf11
opened
2 years ago
7
How to change the batch_size of DETR3D when training?
#29
Bosszhe
closed
2 years ago
4
How to visualize the results of predictions on images?
#28
kaixinbear
closed
2 years ago
2
feature_sampling problem
#27
Yvaine
closed
2 years ago
3
Strange phenomenon on the KITTI dataset.
#26
lianqing11
opened
2 years ago
4
Why not use camera intrinsics when converting the point cloud coordinate system to the image coordinate system?
#25
mayechi
opened
2 years ago
3
多相机环视3D目标检测微信交流群
#24
chaytonmin
opened
2 years ago
5
TypeError: can't pickle dict_keys objects
#23
Egozjuer
opened
2 years ago
3
Next