[HN Gopher] Show HN: Real-Time YOLO Object Detection in Elixir: ...
       ___________________________________________________________________
        
       Show HN: Real-Time YOLO Object Detection in Elixir: Fast, Simple,
       Extensible
        
       Author : alvises
       Score  : 41 points
       Date   : 2024-12-06 17:39 UTC (5 hours ago)
        
 (HTM) web link (github.com)
 (TXT) w3m dump (github.com)
        
       | steinvakt2 wrote:
       | How is this different/better than ultralytics? I've used
       | ultralytics extensively and it allows training and testing using
       | roughly 5 lines of code.
        
         | alvises wrote:
         | The main difference is that this library is built for Elixir,
         | which means you can take advantage of the Elixir real-time and
         | distributed capabilities to easily integrate YOLO into
         | scalable, fault-tolerant systems.
        
         | andy_ppp wrote:
         | It actually uses ultralytics version 8 (but that seems to be on
         | version 11!) so it's a bit behind but should have excellent
         | performance still.
         | 
         | It mentions allowing you to use additionally trained model is
         | an enhancement that is being worked on for the future!
         | 
         | Still, I think it's fantastic that you can take a great model
         | in python and make it available in the Elixir machine learning
         | framework, allowing Elixir programs to integrate with this type
         | of models at a low level without completely reinventing all the
         | effort to train and build them in the first place!
        
           | alvises wrote:
           | Yeah, v11 has been released recently.
           | 
           | Totally agree about ONNX, without it I wouldn't probably even
           | started making this library. However, something I want to try
           | is to code the network with Nx/Axon, without using onnx. The
           | main advantage I see moving from onnx to Axon is to take
           | advantage of all the new cool stuff like Nx sharding:
           | https://dockyard.com/blog/2024/11/06/2024/nx-sharding-
           | update...
        
         | inquisitive-me wrote:
         | Do you actually use Ultralytics for a product? I don't
         | understand why people would want to depend on that mess. There
         | is a terrible license for anything commercial And from what
         | I've heard the code is a mess so doing any customization for
         | your specific use case is a nightmare
        
           | alvises wrote:
           | It's not built around ultralytics model and it doesn't use
           | its code, just del model and weights as a demo. The library
           | is meant to be extensible and to run custom yolo models soon!
        
             | inquisitive-me wrote:
             | Awesome glad to hear that. I was replying to the comment
             | cause I think anything that is not Ultralytics and is truly
             | open source is a good thing. And I was surprised to here
             | someone actually saying anything good about Ultralytics
        
         | ImageXav wrote:
         | I've found that YOLOv7 [0] tends to perform better across the
         | board than anything ultralytics has produced, without the
         | horrendous licensing.
         | 
         | [0] https://github.com/WongKinYiu/yolov7
        
           | alvises wrote:
           | That's great! I'm going to support it then :)
        
       | agarwa90 wrote:
       | This is amazing!
        
       ___________________________________________________________________
       (page generated 2024-12-06 23:00 UTC)