Complex-YOLOv3

Complex-YOLOv3 copied to clipboard

zhouzhou33

Here is a Wechat group for discussing the lidar-based perception technologies. Welcome to join us.

4


deepmeng

How to test on my own lidar data?

1

Anyone can help me? @ghimiredhikura

chenxyyy

Accuracy and Performance

19

@ghimiredhikura its an amazing work you have done here , i had few queries 1. What is the mAP of ur model 2. since yolov3 is preferred for faster inference...

abhigoku10

How to choose anchor boxes in complex_yolov3.cfg?

6

Anchor boxes are choosen in complex_yolov3.cfg anchors = 11,14,-3.14, 11,14,0, 11,14,3.14, 11,25,-3.14, 11,25,0, 11,25,3.14, 23,51,-3.14, 23,51,0, 23,51,3.14 Is there is any basis on the top of which anchors are designed....

chowkamlee81

Why change x and y position in function build_yolo_target?

3

First, I noticed we need to permute `x` and `y` label in `build_yolo_target` in [kitti_bev_utils.py,](https://github.com/ghimiredhikura/Complex-YOLOv3/blob/master/utils/kitti_bev_utils.py#L125) which confuses me. ``` one_target.append(cl) one_target.append(y1) one_target.append(x1) ``` Besides, in [test_detection.py,](https://github.com/ghimiredhikura/Complex-YOLOv3/blob/master/test_detection.py#L152), `w` and `l` need...

usherbob

architecture modifications

1

@ghimiredhikura do u have any thoughts of changing the yolov3 architecture to obtain better accuracy for pedestrian and cyclist classes

abhigoku10

Module doesn't work if BEV RGB images sizes are not square?

5

I need to change FOV from [0,50] and [-25,25] to [0,40] and [-40,40]. In trhis case iam assuming BEV RGB images are of size 480*960, then YOLO modules crashes. KIndly...

chowkamlee81

Trained model

1

@ghimiredhikura Is it possible to share your trained model ???

abhigoku10

I noticed my RGB maps were oddly split down the middle and flipped. It's due to the negative indices rolling over. A fix would involve adding `+ Height / 2`...

UsmannK