CVApr 8, 2018

YOLOv3: An Incremental Improvement

arXiv:1804.02767v125238 citations
Originality Synthesis-oriented
AI Analysis

This work provides faster object detection for computer vision applications, but it is incremental.

The authors improved YOLO with incremental design changes, achieving 57.9 mAP@50 at 51 ms on a Titan X, which is similar in accuracy to RetinaNet but 3.8 times faster.

We present some updates to YOLO! We made a bunch of little design changes to make it better. We also trained this new network that's pretty swell. It's a little bigger than last time but more accurate. It's still fast though, don't worry. At 320x320 YOLOv3 runs in 22 ms at 28.2 mAP, as accurate as SSD but three times faster. When we look at the old .5 IOU mAP detection metric YOLOv3 is quite good. It achieves 57.9 mAP@50 in 51 ms on a Titan X, compared to 57.5 mAP@50 in 198 ms by RetinaNet, similar performance but 3.8x faster. As always, all the code is online at https://pjreddie.com/yolo/

Code Implementations311 repos

Data from Papers with Code (CC-BY-SA-4.0)

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes