arnasdev / arnasdev/Unity-YOLO2-Object-Detection

Not the same objects detected

Open
#4 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
C#
Stars
6
Forks
1
PR merge metrics
No merged PRs in 30d

Description

Hi,
When I run this project on Unity2018 I see differences on the objects detected compared to the project you mentioned [[C#] OpenCvSharp DNN with YOLO2]( https://www.died.tw/2018/01/c-opencvsharp-dnn-with-yolo2.html ).

![rgertg](https://user-images.githubusercontent.com/33813291/56356361-f30c2380-61d8-11e9-98b7-a859db7c5612.PNG)

Every time I run the project on unity it doesn't show me the cycle detected.

But here [[C#] OpenCvSharp DNN with YOLO2]( https://www.died.tw/2018/01/c-opencvsharp-dnn-with-yolo2.html ) insted yes.

![fbbhn](https://user-images.githubusercontent.com/33813291/56356779-14b9da80-61da-11e9-9cc4-aa3eeabafe9c.PNG)

This problem is repeated with all the other images, it seems as if it cannot identify the same number of objects. I also tried to use the same "yolo-voc.weights" model and the same "yolo-voc.cfg" weights provided in the other project, but it remains the same.

Is there any solution? What I am doing wrong?

Thanks a lot.

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by comparing the Unity 2018 project output with the linked OpenCvSharp DNN with YOLO2 reference, using the mentioned yolo-voc.weights and yolo-voc.cfg files. Check why the Unity run detects fewer objects, including the missing cycle, and reproduce the discrepancy on the supplied images. Done means the same objects are detected consistently in both projects.

Written by the indexing model from the issue text.

Assessment

Tech stack
csharp, unity
Domain
computer-vision, game-dev
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.