YOLOv6: Redefine state-of-the-art for object detection

Re:YOLOv6: Redefine state-of-the-art for object detection by echelon on Wednesday June 29, 2022 @05:40AM (#31918074)

At this point so many different people have reused "YOLOv{n}" that the lineage is broken. I don't know how or why it happened, but it continues to be a thing.

[ Parent | on HN ]

Re:YOLOv6: Redefine state-of-the-art for object detection by chriskanan on Wednesday June 29, 2022 @05:49AM (#31918129)

I assume this happened because Joseph Redmon, the creator of v1 - v3, left the field of computer vision [1]. So then other people just took (stole?) the name.

[1]. https://syncedreview.com/2020/02/24/yolo-creator-says-he-sto...

[ Parent | on HN ]

Re:YOLOv6: Redefine state-of-the-art for object detection by basedbertram on Wednesday June 29, 2022 @06:43AM (#31918446)

At the very least v4 has a paper attached to it. The authors of v5 have claimed that they'll release a paper at some point, but it has never materialized. It doesn't seem like v6 even makes that promise.

[ Parent | on HN ]

Re:YOLOv6: Redefine state-of-the-art for object detection by npRandom on Wednesday June 29, 2022 @07:20AM (#31918676)

Yup. They don't plan to release a paper but a technical report https://github.com/meituan/YOLOv6/issues/95

[ Parent | on HN ]