NeilNie/speed_estimation
Using deep learning to predict the speed of a moving vehicle.
Leverages DeepMind's Inflated 3D ConvNet (i3D) architecture to extract spatial-temporal features from video sequences, enabling regression of vehicle speed beyond single-frame prediction. The model inflates 2D ImageNet-pretrained filters into 3D convolutions for seamless video understanding, with final layers adapted for continuous speed output. Includes visualization techniques demonstrating learned traffic behaviors like stopping and yielding patterns.
No commits in the last 6 months.
Stars
11
Forks
4
Language
Python
License
MIT
Category
Last pushed
Sep 15, 2018
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/NeilNie/speed_estimation"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
arasgungore/car-brand-recognition
Final project assigned for the Introduction to Image Processing (EE 475) course in the Spring...
TariqAlhathloul/Smart-Street
detect and classify road lines, enhancing road safety by identifying no-overtaking zones.
CWLeonis/Car_detection_Deep_learning
detecting model and the name of the cars with deep neural networks like VGG-16 , YOLOv5 and YOLOv8
NaimurRahman-Niaz02/Car-Classification-and-Image-Processing
A Flask-based web application that combines deep learning and image processing to classify car...
tmridwan/vehicle_classifier
detect every types of vehicle