# Object Detection in Julia

**URL:** <https://discourse.julialang.org/t/object-detection-in-julia/65115>\
**Category:** Performance\
**Created:** [July 22, 2021, 3:44pm UTC](https://discourse.julialang.org/t/object-detection-in-julia/65115 "2021-07-22T15:44:25Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![ppkvm](https://avatars.discourse-cdn.com/v4/letter/p/fbc32d/32.png) [@ppkvm](https://discourse.julialang.org/u/ppkvm)\
**Post date:** [July 22, 2021, 3:44pm UTC](https://discourse.julialang.org/t/object-detection-in-julia/65115/1 "2021-07-22T15:44:25Z")

</div>

I would like to see examples of Object Detection using YOLO and SSD implemented in Julia.  
Has anyone tried this before?

---

<div class="post-metadata">

**Author:** ![jling](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/jling/32/212909_2.png) [@jling](https://discourse.julialang.org/u/jling)\
**Post date:** [July 22, 2021, 3:52pm UTC](https://discourse.julialang.org/t/object-detection-in-julia/65115/2 "2021-07-22T15:52:37Z")

</div>

[https://github.com/IanButterworth/YOLO.jl](https://github.com/IanButterworth/YOLO.jl)

---

<div class="post-metadata">

**Author:** ![ianshmean](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/ianshmean/32/216042_2.png) [@ianshmean](https://discourse.julialang.org/u/ianshmean)\
**Post date:** [July 22, 2021, 3:59pm UTC](https://discourse.julialang.org/t/object-detection-in-julia/65115/3 "2021-07-22T15:59:43Z")

</div>

YOLO.jl has a knet backend.

ObjectDetector is a bit more featured, with a Flux backend.  
[https://github.com/r3tex/ObjectDetector.jl](https://github.com/r3tex/ObjectDetector.jl)

There’s also a wrapper of AlexeyAB’s fork of Darknet  
[https://github.com/IanButterworth/Darknet.jl](https://github.com/IanButterworth/Darknet.jl)

---

<div class="post-metadata">

**Author:** ![jling](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/jling/32/212909_2.png) [@jling](https://discourse.julialang.org/u/jling)\
**Post date:** [July 22, 2021, 4:00pm UTC](https://discourse.julialang.org/t/object-detection-in-julia/65115/4 "2021-07-22T16:00:51Z")

</div>

I linked to your YOLO partly because your README has pointers to other related projects

---

<div class="post-metadata">

**Author:** ![ppkvm](https://avatars.discourse-cdn.com/v4/letter/p/fbc32d/32.png) [@ppkvm](https://discourse.julialang.org/u/ppkvm)\
**Post date:** [July 22, 2021, 4:25pm UTC](https://discourse.julialang.org/t/object-detection-in-julia/65115/5 "2021-07-22T16:25:54Z")

</div>

Great, this helps.

---

<div class="post-metadata">

**Author:** ![ppkvm](https://avatars.discourse-cdn.com/v4/letter/p/fbc32d/32.png) [@ppkvm](https://discourse.julialang.org/u/ppkvm)\
**Post date:** [July 22, 2021, 4:27pm UTC](https://discourse.julialang.org/t/object-detection-in-julia/65115/6 "2021-07-22T16:27:16Z")

</div>

Also the performance, what is the best way to measure the performance in terms of memory footprint and training and inference time ?

---

<div class="post-metadata">

**Author:** ![Iulian.Cioarca](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/iulian.cioarca/32/30166_2.png) [@Iulian.Cioarca](https://discourse.julialang.org/u/Iulian.Cioarca)\
**Post date:** [August 17, 2021, 10:17am UTC](https://discourse.julialang.org/t/object-detection-in-julia/65115/7 "2021-08-17T10:17:35Z")

</div>

Bumping the topic a bit. I also found: [https://github.com/DhairyaLGandhi/maskrcnn](https://github.com/DhairyaLGandhi/maskrcnn)  
Does any of them have the training capability as well? At least to un-freeze the last layers because I would need to re-train them on a custom dataset.