AI Model for Vehicle Surface Damage Detection – Scratch, Dent & Blur Identification -- 2

Title: AI Developer for Surface Defect Detection in Automotive (Image & Video-based)

Description:
We are looking to hire an experienced AI/ML developer to build an intelligent surface defect detection system for automotive applications. The system will identify defects such as dents, scratches, and cracks on a car's exterior from multiple images initially, and later from 360° videos.

Project Scope (Phase 1):

Input: 8 or more images capturing the full exterior of a vehicle


Process: Submit the images to an API to be developed by you


Output:


Annotated images showing any detected defects


Classification of defect type (e.g., dent, scratch, crack)


Identification of defect location (e.g., left door, right door, front bumper, rear panel, etc.)



Project Scope (Phase 2 - Future Milestone):
Input: A 360° video (captured manually by hand using a phone)


Output: Same as Phase 1 — annotated frames or video with defect type and location



Deliverables:
API endpoint to receive and process image input


Annotated image output with defect markings and labels


JSON metadata indicating:


Type of defect


Severity level (if feasible)


Location of defect on vehicle


Documentation for deployment and usage



Important Questions (to be answered in your proposal):
Experience:
Do you have prior experience with surface defect detection, object detection, or automotive vision systems? Please provide relevant project links or case studies.


Model Choice:
Which open-source model(s) will you be using for this task? (e.g., YOLOv8, Detectron2, SAM, etc.)
Please justify your choice and provide links to the model(s).


Infrastructure Requirements:
What infrastructure (GPU/CPU, RAM, storage, cloud provider preference, etc.) will be needed to run and host this model efficiently?


Data Requirements:
What data do you need from us to start working on this project? (e.g., labeled/unlabeled images, video samples, defect taxonomy, etc.)



Preferred Skills:
Computer Vision (CV) and Deep Learning


Object Detection and Segmentation (YOLO, Mask R-CNN, etc.)


OpenCV, PyTorch, TensorFlow


API development (Flask/FastAPI)


Experience with cloud deployment (AWS/GCP/Azure)



Additional Notes:
This project may evolve into a long-term engagement including Phase 2.


Accuracy and performance will be key. Prior work in similar automotive/inspection domains will be highly preferred.


*****************************************************************************************************

Please start your proposal with “Defect Detection Project – [Your Name]” so we know you read the description fully.

*****************************************************************************************************