-
Hello,
Thanks for this great software package. It has been extremely easy to use and performs great.
For my specific application (light sheet microscopy), I deduced that it may be beneficial to …
-
Hello,
I apply function `diff_iou_rotated_3d` from mmcv.ops to implement 3D IoU loss.
However, the computed IoU can be larger than 1 in some circumstances, which is wrong by definition.
But if …
-
# Hyper-parameters
1. Model Type : UNet, 티라미수 ~~먹고싶다~~
2. Model Layer 수 : 20 layer => 30 layer => model capacity에 기여
3. Upsample : bilinear vs transposed convolution => model capacity에 기여
4…
-
## 🚀 Feature
It would be great if the backward pass of the rotated IoU CUDA module would be implemented, since this way one could train a detector to directly optimize it.
Moreover, recent papers i…
-
## 🚀 Feature
A loss functions API in torchvision.
## Motivation
The request is simple, we have loss functions available in torchvision
E.g. `sigmoid_focal_loss` , `l1_loss`. But these are qu…
-
## 튜토리얼에 있는 데이터셋의 구조
---
- `/content/object_detection_demo/data/images/test` 에 있는 10.xml
![image](https://user-images.githubusercontent.com/30471027/59108733-962a1f80-8976-11e9-8e94-df3b8dbfe22b.…
-
Hi all, I am training the network using COCO pre-trained weights but the results are way lower than expected. The dataset I am using is the COCO dataset 2017. I have used different configurations and …
-
i don't understand exactly approximate joint training method.
i know RPN and detector merged as a one network during training.
the forward path is started pre trained conv network and pass from RPN …
-
## 🚀 Feature
Implement YOLO v5 from `torch.hub`.
This library removes such `dataset` abstraction and aims to provides a clean modular interface to models.
Some key points to note: -
1. YOLO Ac…
-
Implement Yolo-LSTM detection network that will be trained on Video-frames for mAP increasing and solve blinking issues.
* https://arxiv.org/abs/1705.06368v3
* https://arxiv.org/abs/1506.04214v2
…