10. Object Detection Lab
This optional lab exercise is an opportunity to practice rapid object detection suitable for implementation in autonomous vehicles. The current lab is focused on a general understanding of used SSD for object detection, as applied to detecting other vehicles from a driving video. The proficiency and understanding developed in this lab can ultimately be used to detect other relevant objects, such as Traffic lights.
In lab this you will:
- Learn about MobileNets and separable depthwise convolutions.
- The SSD (Single Shot Detection) architecture used for object detection
- Use pretrained TensorFlow object detection inference models to detect objects
- Use different architectures and weigh the tradeoffs.
- Apply an object detection pipeline to a video.
Clone the GitHub repository at https://github.com/udacity/CarND-Object-Detection-Lab , open the notebook and work through it!
Requirements
Install environment with Anaconda :
conda env create -f environment.yml
Change TensorFlow pip installation from
tensorflow-gpu
to
tensorflow
if you don't have a GPU available.
The environment should be listed via
conda info --envs
:
# conda environments:
#
carnd-advdl-odlab /usr/local/anaconda3/envs/carnd-advdl-odlab
root * /usr/local/anaconda3
Further documentation on working with Anaconda environments .
Particularly useful sections:
https://conda.io/docs/using/envs.html#change-environments-activate-deactivate
https://conda.io/docs/using/envs.html#remove-an-environment
Resources
- TensorFlow object detection model zoo
- Driving video