TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

YOLOv4: Optimal Speed and Accuracy of Object Detection

234 pointsby groarabout 5 years ago

7 comments

rgovostesabout 5 years ago
The original author of YOLO stopped working on it[1]. Alexey Bochkovskiy, aka AlexeyAB, created a fork on GitHub and wrote an extensive guide to customizing YOLO&#x27;s network architecture, added new features, and has answered zillions of questions.<p>1: <a href="https:&#x2F;&#x2F;twitter.com&#x2F;pjreddie&#x2F;status&#x2F;1230524770350817280" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;pjreddie&#x2F;status&#x2F;1230524770350817280</a>
评论 #22995950 未加载
评论 #22997035 未加载
评论 #22999728 未加载
评论 #22998750 未加载
评论 #22995808 未加载
punnerudabout 5 years ago
And code on Github: <a href="https:&#x2F;&#x2F;github.com&#x2F;AlexeyAB&#x2F;darknet" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;AlexeyAB&#x2F;darknet</a><p>PyTorch version: <a href="https:&#x2F;&#x2F;github.com&#x2F;Tianxiaomo&#x2F;pytorch-YOLOv4" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Tianxiaomo&#x2F;pytorch-YOLOv4</a>
评论 #23002429 未加载
gjsteinabout 5 years ago
I object to the use of the word &quot;optimal&quot; for a task like object detection; it feels counterproductive to claim that this is the &quot;optimal&quot; way of solving such a broad and complex problem. Great results, but their language needs some tempering.
评论 #23012858 未加载
评论 #22997959 未加载
Reigngt09about 5 years ago
A video on YOLOv4 - Really informative. <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=_JzOFWx1vZg" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=_JzOFWx1vZg</a>
mchusmaabout 5 years ago
I have had a lot of fun working with YOLO v3 for robotics applications, very excited to try these updates. Thanks to the authors for the updates and good documentation. Good object recognition is the backbone of a huge range of future applications, and YOLO has been a good option for a while.
SethTroabout 5 years ago
I&#x27;m a little skeptically of the Swish implementation after looking at Table 2.<p>Method | Top 1 | Top 5 No-op | 78% | 94% Swish | 64.5% | 86% Mish | 79% | 94.5%<p>Swish is the only value that decreases performance (and by a huge magnitude) but a very related methodology improves performance hummm...
Reigngt09about 5 years ago
<a href="https:&#x2F;&#x2F;medium.com&#x2F;@riteshkanjee&#x2F;yolov4-superior-faster-more-accurate-object-detection-7e8194bf1872" rel="nofollow">https:&#x2F;&#x2F;medium.com&#x2F;@riteshkanjee&#x2F;yolov4-superior-faster-more...</a><p>Article on YoloV4