Offline Windows Video Frame Analyzer
Employer not named by the sourceRemote
Frontier is not the employer and does not collect applications.
About this role
Python, Windows Desktop, C# Programming, Software Architecture, Machine Learning (ML), C++ Programming, OpenCV, Object Detection · I need a Windows-only desktop application that runs entirely offline. The workflow is straightforward:
1. I open the program, browse my local folders, and pick any .mp4 file. 2. Before the scan starts, I set the frame-sampling rate—this could be every frame, every nth frame, or even a custom decimal rate such as 2.5 fps. 3. The app then performs object detection on those chosen frames. I do not mind which framework you use (YOLO, TensorFlow, OpenCV, ONNX, etc.) as long as everything is bundled so the tool works without an internet connection. 4. When the run finishes, the program produces a plain .txt sheet: one line per analysed frame containing the timestamp and the symbols / class names it detected. No other file formats are required.
Deliverables • Stand-alone Windows executable (installer or portable) • All necessary model files packaged for offline use • Source code and brief build/run instructions
Acceptance Criteria • User can freely select any local .mp4 file. • Frame-sampling rate is fully configurable for each run. • Object detection runs offline and lists detections accurately in the .txt sheet. • Output file is generated automatically in th