Project Description

Senior Computer Vision Engineer – Logo Detection & Tracking in Video
Location: Remote
Type: Contract / Project-based
Experience Level: Senior / Expert

We are seeking a Senior Computer Vision Engineer to build a high-performance engine for detecting and tracking brand logos in professional event videos. This is not a simple cat/dog classification task—we require robust detection under challenging conditions.

Project Overview
The system must take:

Reference Logo (Image)
Target Video
…and output a detailed report including:

Timestamp Logs: Every second/frame where the logo is visible.
Spatial Analytics: Precise placement and coordinates of the logo (e.g., helmet, bib, banner).
Robust Detection: Handle non-rigid deformations (wrinkled clothes), extreme perspective shifts, and high-speed motion blur.

Technical Requirements
Zero-Shot Detection / Feature Matching:Must work instantly for new logos without retraining.
Experience with LightGlue, SuperPoint, LoFTR, or YOLO-World.
Video Pipeline Expertise:Efficient frame processing using OpenCV, FFmpeg, CUDA, or similar.
Geometric Invariance:Ability to solve perspective and affine transformations.
Robust Tracking:Implement temporal tracking with DeepSORT, ByteTrack, or similar to maintain identity across pans and zooms.

Who We’re Looking For
Proven experience building production-level CV systems.
Ability to optimize for high accuracy (80%+) in real-world videos.
Skilled at handling distorted, tilted, or moving objects.
Comfortable designing end-to-end pipelines from input video to detailed analytics.

Deliverables
Working prototype capable of detecting and tracking new logos in unseen videos.
Timestamped and spatially annotated output.
Optimized and maintainable code suitable for production deployment.

How to Apply
Please include:

A demo or description of a project where you successfully detected and tracked distorted, tilted, or high-speed objects.
Your approach to zero-shot logo detection and tracking.
Any prior experience in sports or event video analytics (optional but preferred).

This is a high-impact project aimed at production-grade deployment. Only serious candidates with strong CV/ML experience and video pipeline expertise should apply.