Giters
megvii-research
/
AnchorDETR
An official implementation of the Anchor DETR.
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
326
Watchers:
11
Issues:
48
Forks:
34
megvii-research/AnchorDETR Issues
载入预训练模型发现attention 维度不对,可能的问题是 ?
Closed
4 months ago
Comments count
4
Multiple Predictions for Each Anchor Point
Closed
6 months ago
Comments count
1
How to train one's own dataset
Closed
6 months ago
Comments count
8
Question about calculating GFLOPs
Closed
a year ago
Comments count
2
Is there any way to get the coordinates(polygon shape returns)?
Closed
a year ago
Comments count
4
How to replace resnet50 with your own backbone
Closed
a year ago
Comments count
2
bs, l, c, h, w = srcs.shape
Closed
2 years ago
Comments count
1
Why GroupNorm in self.input_proj ?
Closed
2 years ago
Comments count
2
Some questions about reference point and object query?
Closed
2 years ago
Comments count
6
The comparison with DETR
Closed
2 years ago
Comments count
4
测试问题
Closed
2 years ago
Comments count
6
Does the model always output 100 detection results on one image?
Closed
2 years ago
Comments count
2
关于feature shape的问题
Closed
2 years ago
Comments count
2
Hello! The Feature position in figure3 is the positional encoding in DETR
Closed
2 years ago
Comments count
2
Hello the default model is AnchorDETR-C5 or AnchorDETR-DC5 ?
Closed
2 years ago
Comments count
2
Why not input updated reference points to next layer decoder?
Closed
2 years ago
Comments count
2
Trying to understand my results on a custom dataset
Closed
2 years ago
Comments count
9
Details about GFLOPs
Closed
2 years ago
Comments count
3
Points visualization
Closed
2 years ago
Comments count
5
num_feature_levels > 1
Closed
2 years ago
Comments count
43
question about extended to panoptic segmentation
Closed
2 years ago
Comments count
21
NAN error: assert (boxes1[:, 2:] >= boxes1[:, :2]).all()
Closed
2 years ago
Comments count
10
Model can not achieve a reasonable performance
Closed
2 years ago
Comments count
8
About batch size and lr
Closed
2 years ago
Comments count
3
Pretrained weights transfer to d2 evaluation result not same
Closed
2 years ago
Comments count
3
something wrong with evaluating
Closed
2 years ago
Comments count
2
About Visualization
Closed
2 years ago
Comments count
3
Code for PointDETR
Closed
2 years ago
Comments count
1
coco test-dev
Closed
3 years ago
Comments count
4
points visualization
Closed
2 years ago
Load pretrained model AnchorDETR-DC5.
Closed
2 years ago
Comments count
2
How extract precision, recall and f1-score metrics
Closed
2 years ago
Comments count
5
Downloaded AnchorDETR R50 C5 weights evaluation only got 38.5
Closed
2 years ago
Comments count
3
How to get the 151 GFLOPs?
Closed
3 years ago
Comments count
2
AP stops at 30 at iteration 250000
Closed
3 years ago
Comments count
1
关于在自己数据集上的应用
Closed
3 years ago
Comments count
1
GPU memory consumption of AnchorDETR compared to DeformableDETR
Closed
3 years ago
Comments count
2
Overfitting when loading the pretrained model to train own dataset
Closed
3 years ago
Comments count
15
why the output is 900 while DETR is 100
Closed
3 years ago
Comments count
3
onnx model can not be simpiflied and pass onnx.check and wierd output
Closed
3 years ago
Comments count
10
How many encoder and decoder layers used in AnchorDETR?
Closed
3 years ago
Comments count
30
Pretrained model
Closed
3 years ago
Comments count
3
Your module has parameters that were not used in producing a loss.
Closed
3 years ago
Comments count
3
Performance of the baseline
Closed
3 years ago
Comments count
1
how to use custom data
Closed
3 years ago
Comments count
3
Question about the time cost per epoch
Closed
3 years ago
Comments count
4
How to finetune AnchorDETR on custom dataset?
Closed
3 years ago
Comments count
2
Some questions about your code
Closed
3 years ago
Comments count
7