https://github.com/experiencor/image-to-3d-bbox

Build a CNN network to predict 3D bounding box of car from 2D image.
https://github.com/experiencor/image-to-3d-bbox

bounding-boxes cnn-keras convolutional-neural-networks deep-learning kitti-dataset regression self-driving-car

Last synced: 5 months ago
JSON representation

Build a CNN network to predict 3D bounding box of car from 2D image.

• Host: GitHub
• URL: https://github.com/experiencor/image-to-3d-bbox
• Owner: experiencor
• Created: 2017-07-10T09:07:47.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2019-07-09T02:36:29.000Z (over 6 years ago)
• Last Synced: 2025-03-31T11:41:34.379Z (6 months ago)
• Topics: bounding-boxes, cnn-keras, convolutional-neural-networks, deep-learning, kitti-dataset, regression, self-driving-car
• Language: Jupyter Notebook
• Homepage: https://experiencor.github.io/sdc_3d.html
• Size: 688 KB
• Stars: 240
• Watchers: 7
• Forks: 58
• Open Issues: 8
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
This is an implementation in Keras of the paper "3D Bounding Box Estimation Using Deep Learning and Geometry" (https://arxiv.org/abs/1612.00496).

# Demo on a street video

![Foo](https://j.gifs.com/zmWKjO.gif)


# Usage

+ "3D Box Regression.ipynb" => to construct and train the network to regress 3D bounding boxes.

+ "Final KITTI Evaluation.ipynb" => to draw 3D bounding boxes on 2D images.