Object detection using machine learning techniques
Computer Vision is now an active research area under Deep Learning and Object Detection is an integral part of Computer Vision. Having a more complete idea of Deep Learning-based Object Detection Application, I start building up my own object detector for 4 classes of objects: person, car, bicycle...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/78281 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Computer Vision is now an active research area under Deep Learning and Object Detection is an integral part of Computer Vision.
Having a more complete idea of Deep Learning-based Object Detection Application, I start building up my own object detector for 4 classes of objects: person, car, bicycle and motorbike. This object detector is tailored for use in Singapore and I achieve a good mAP of 42% which is much higher than the official statistics provided by TensorFlow. In addition, I not only conduct inference test on my object detector model with image inputs, but also I test it with a live stream video by using Webcam. What’s more interesting is that I deploy this object detection model in a mobile device. It can simulate the driving scenario by realizing real-time object detection. |
|---|