VisionLearningGroup / DA_Detection

Implementation of "Strong-Weak Distribution Alignment for Adaptive Object Detection"

http://cs-people.bu.edu/keisaito/research/CVPR2019.html

VisionLearningGroup/DA_Detection Issues

ImportError: No module named cython_bbox
Updated a year ago1
How to draw the picture like this in your paper?
Updated 2 years ago
some problem about train from cityscape to foggy cityscape
Updated 2 years ago17
RuntimeError: size mismatch occurs when using `--lc --gc` for `trainval_net_global_local.py`
Closed 6 years ago5
pretrained model load error
Updated 2 years ago
domain label are different from paper
Updated 2 years ago2
about the focalloss
Updated 2 years ago
DA_faster evaluate on clipart score is higher than your papers
Updated 3 years ago
where to change the threshold
Updated 3 years ago
import error when train
Updated 3 years ago1
Data loading problem
Updated 3 years ago
Providing Pre-trained model for sim10 -> City. ?
Updated 3 years ago1
The AP is slower than the paper report,I don't know what caused this.
Updated 4 years ago1
how to implement custom class?
Closed 4 years ago
About global alignment loss and local alignment loss
Updated 4 years ago
Epochs trained for cityscapes to foggy cityscapes DA
Updated 4 years ago1
The dimensions of the parameters of the trained and tested models do not match
Updated 4 years ago3
ModuleNotFoundError: No module named 'lib.model.roi_crop._ext.roi_crop._roi_crop'
Updated 4 years ago4
Once trained with source and target domain data, can we generate a model,wherein model will predict new dataset/images similar to target domain dataset.
Closed 4 years ago
Unexpected key(s) in state_dict: "netD.bn1.num_batches_tracked", "netD.bn2.num_batches_tracked", "netD.bn3.num_batches_tracked".
Updated 4 years ago
cityscape.py did not filter the target domain images
Updated 4 years ago2
Few questions for reproducing the results on Watercolor
Updated 5 years ago2
About CycleGAN settings
Updated 5 years ago
Visualization of features obtained
Updated 5 years ago
How can I see the baseline results?
Updated 5 years ago
why calculated sigmoid twice？
Closed 5 years ago
Multi Gpu training issue
Updated 5 years ago2
Try this on SSD
Updated 5 years ago
lower result than paper
Updated 5 years ago6
The reason why using all the clipart data set when adapt from VOC to clipart?
Closed 5 years ago17
Ablation study for w/o local context?
Updated 5 years ago
size mismatch when loading the trained checkpoint
Updated 5 years ago5
Hi~I have a question. Foggy Cityscapes dataset has multiple indexes (0.01, 0.02, 0.005).Which index is used in this paper
Updated 5 years ago1
Results on adpatation from PASCAL VOC to Clipart Dataset does not match
Updated 5 years ago
RuntimeError: cuDNN version mismatch: PyTorch was compiled against 7102 but linked against 7600
Closed 5 years ago1
How to get the dataset Cityscapes_car?
Updated 5 years ago
Can this project run with multi-gpus?
Updated 5 years ago
Warning: NaN or Inf found in input tensor.
Closed 5 years ago4
foggy Cityscapes preparation
Closed 6 years ago4
Running Error
Updated 5 years ago3
No module named 'PIL', 'torchvision'
Updated 5 years ago1
Data preparation
Closed 6 years ago1
About Backbone
Updated 6 years ago1
Error in Clipart Dataset
Closed 6 years ago1
DA_Detection/lib/model/nms/src/nms_cuda.c: No such file or directory
Updated 6 years ago2
ImportError: torch.utils.ffi is deprecated. Please use cpp extensions instead.
Updated 6 years ago2