ethanhe42 / softer-NMS

Bounding Box Regression with Uncertainty for Accurate Object Detection (CVPR'19)

https://github.com/ethanhe42/KL-Loss

ethanhe42/softer-NMS Issues

The final ideal goal of learned standard deviation？
Updated 3 years ago
how could I train my own dataset use your repo?
Closed 6 years ago7
我在fine turn我之前训练的模型的时候，为什么bbox_pred_std_abs_logw_loss为负值？
Closed 6 years ago3
Final loss when you train your model
Closed 5 years ago2
How does the loss function train the network?
Closed 4 years ago1
how to implement softer-NMS by pytorch version
Closed 4 years ago4
could you show the code position of softNMS and variance-voting?
Closed 4 years ago2
use for YOLO
Closed 4 years ago1
Can this method be used in one-stage detector like retinanet?
Closed 4 years ago1
第二阶段训练后，mAP值降低了？？
Closed 5 years ago3
when train on our own dataset, error occurs,
Closed 5 years ago4
x1y1x2y2 Coordinates only in FasterRCNN head?
Closed 5 years ago1
Detectron module
Closed 5 years ago2
KL loss
Closed 5 years ago1
KL loss
Closed 6 years ago1
Some question about "confs"
Closed 6 years ago3
How can softer-NMS be added to its own network structure?
Closed 6 years ago1
what are the differences between your softer-NMS detectron version and original detectron?
Closed 6 years ago2
How can softer-NMS be added to its own network structure?
Closed 6 years ago1