vg11072001 / car_damage_detection Public

Notifications You must be signed in to change notification settings
Fork 0
Star 3

• Use of TensorFlow Object Detection API to train Custom Object Detector. • Around 6000 car images data used to train the model and annotated in 3 parts replacement, dent and scratch. • Completed in steps Data Annotation & Cleaning, Object detection models, Tf Object Detection API and Preprocessing the data

3 stars 0 forks Branches Tags Activity

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 12 Commits
models		models
scripts/preprocessing		scripts/preprocessing
workspace/traning_demo		workspace/traning_demo
README.md		README.md
car_damage_detection_1.ipynb		car_damage_detection_1.ipynb

Repository files navigation

Car Damage Detection

Dataset

Used 6000 images of damaged cars. Overview -

Dataset Anotation

Dataset divided into 3 categories:

Replacement
Scratch
Dent

Used labelImg to Annotate

Inference

Plots

About

• Use of TensorFlow Object Detection API to train Custom Object Detector. • Around 6000 car images data used to train the model and annotated in 3 parts replacement, dent and scratch. • Completed in steps Data Annotation & Cleaning, Object detection models, Tf Object Detection API and Preprocessing the data

python machine-learning deep-learning tensorflow object-detection ssd-mobilenet anotation

Report repository

Releases

No releases published

Packages

No packages published

Languages