Task 1. Background substraction. Segmentation metrics. Precision and recall.
- Run the function evaluateFolder() in main.py
- See the implementation of evaluateFolder() in evaluation.py
Task 2. Background susbtraction. Segmentation metrics. Temporal analysis.
- Run the function evaluateTemporal() in main.py
- See the implementation of evaluateTemporal() in evaluation.py
Task 3. Optical flow evaluation metrics. Mean Squared Error and Percentage of Erroneous Pixels in Non-occluded areas.
- Run the function evaluateOpticalFlow() in main.py
- See the implementation of evaluateOpticalFlow() in optical_flow.py
Task 4. Background substraction. Evaluation of de-synchornized results.
- Run the function dSyncGlobalEvaluation() in main.py to check evolution of overall F1 over de-sync frames
- Run the function dSyncTemporalEvaluation() in main.py to check evolution of F1 per frame over de-sync frames
- See the implementation of dSyncGlobalEvaluation() and dSyncTemporalEvaluation() in evaualtion.py
Task 5. Visual representation of optical flow.
- Run the function plotOpticalFlowHSV() in main.py
- See the implementation details of plotOpticalFlowHSV() in optical_flow.py