YOLO: Real-Time Object Detection

Re:YOLO: Real-Time Object Detection by natvert on Tuesday April 02, 2019 @05:42PM (#19558873)

Interesting this is trending now. We have actually just recently released an improved version of YoloV3 (called G-Darknet) https://github.com/generalized-iou/g-darknet, using GIoU as a loss, which is described here: https://giou.stanford.edu

Also notable in G-Darknet are some tools useful for training (called darkboard), see https://github.com/generalized-iou/g-darknet/tree/master/dar...

[ Parent | on HN ]

Re:YOLO: Real-Time Object Detection by w-m on Wednesday April 03, 2019 @05:55AM (#19562085)

Heads up: the boxes are drawn in the wrong places using Firefox 66 on Ubuntu 18.04. https://imgur.com/a/4d51spv

A bit confusing as the drawn boxes don't match the text. Works with Chromium though.

[ Parent | on HN ]

Re:YOLO: Real-Time Object Detection by natvert on Wednesday April 03, 2019 @09:59AM (#19563336)

Thx, I'll check that

[ Parent | on HN ]

Re:YOLO: Real-Time Object Detection by elephantum on Wednesday April 03, 2019 @06:54AM (#19562316)

Interesting idea!

Though I have a question: in order to calculate C you need a way to attribute proposal and ground truth. It's trivial in case when there's only one instance of each class in the image.

But how does it work, when you're working with a set of same-class object? For example detecting each car in traffic.

[ Parent | on HN ]

Re:YOLO: Real-Time Object Detection by natvert on Wednesday April 03, 2019 @12:29PM (#19564733)

Good question, we use the same method as coco, described in https://arxiv.org/pdf/1405.0312.pdf and implemented in the coco evaluation scripts here: https://github.com/cocodataset/cocoapi/blob/master/PythonAPI... -- basically the best matching proposal, ground truth pair

[ Parent | on HN ]