-
Dear all (@jcohenadad @kennethaweberii ),
Currently, we are using the Dice Coefficient (between manual vs automatic mask) as a metric to assess the performance of newly trained model.
I had a …
-
### What's the feature?
Hi,
I need to run a model on some data (inference or test) and produce an output file in format of COCO json. I saw that in the main branch it is possible using the "--forma…
-
Relevant section from the [CODEX preprint](https://www.biorxiv.org/content/early/2018/02/05/203166):
> A 3D segmentation algorithm was therefore created to combine information from the nuclear stai…
-
### Description
If using centerline (instead of cord segmentations), the resampling, currently done with nearest neighbor, makes the centerline disappear, and the subsequent steps are failing.
w…
-
Thanks, Authors for sharing the codes.
After running these two lines below:
```
python build_batches.py -d Gref -t train
python build_batches.py -d Gref -t val
```
it results in a "train_bat…
-
If we are trying to establish, just the pixels instead of image regions with the highest gradients attached to them, I have heard saliency maps using backpropagation is the best way? Is it possible to…
-
## 🐞 Bug
(Some?) torchvision transforms do not work as padl transforms "out of the box", as implied by the documentation ([Combining transforms](https://lf1-io.github.io/padl/latest/usage/combini…
-
Hi,
Thank you for making this awesome repository! I have a few questions regarding polygon inference:
According to https://github.com/WongKinYiu/yolov7/issues/208 it's possible to input polygons a…
-
## 🐛 Bug
The DGCNN example provided 'examples' cannot achieve the reported accuracy in the paper for ModelNet40 dataset.
I tried to build the same architecture as reported in the paper ( 4 laye…
-
Thanks for your great work! The motion segmentation method you proposed is very novel, but I would like to ask whether the previous mask-rcnn+epipolar distance method in dynamic nerf will cause the mo…