Skip to content

End-to-End Workflow

TrackerKit Workflow

  1. Load video and outputs
  2. Set input video, CSV output, optional rendered video output.
  3. Calibrate detection
  4. Pick detection mode.
  5. Use preview/test detection before full run.
  6. Configure tracking
  7. Set MAX_TARGETS, assignment distance, track lifecycle thresholds.
  8. Run forward/backward tracking
  9. Enable backward pass for better conflict resolution.
  10. Post-process and export
  11. Resolve identities, interpolate gaps as needed.
  12. Save final CSV and optional diagnostics.

PoseKit Workflow

  1. Load image set and project settings.
  2. Label or refine keypoints frame-by-frame.
  3. Use tools (smart select, metadata tags, split generation).
  4. Export/prepare training-ready datasets.

Decision Points That Matter Most

  • Detection mode affects raw input quality to tracker.
  • Reference body size scales multiple heuristics.
  • Post-processing can fix or amplify detection mistakes depending on thresholds.

Failure Pattern Checklist

  • If targets merge often: tighten morphology and assignment distance.
  • If tracks fragment: increase recovery/lost frame thresholds and validate detection confidence.
  • If runtime is slow: reduce resize factor, disable non-critical overlays/histograms, verify GPU backend.

ClassKit Workflow

  1. Create or open a project with source image directories.
  2. Ingest and embed crops using a backbone model.
  3. Cluster embeddings and visualize with UMAP.
  4. Label identity classes manually or via AprilTag auto-labeling.
  5. Train a classification head and evaluate results.
  6. Export labeled datasets for downstream use.

DetectKit Workflow

  1. Curate detection training datasets from TrackerKit exports.
  2. Configure YOLO training parameters.
  3. Launch training and monitor loss curves.
  4. Evaluate model performance on validation sets.

FilterKit Workflow

  1. Load a dataset directory.
  2. Apply filtering criteria (quality, diversity, metadata).
  3. Export the filtered subset for training or analysis.

RefineKit Workflow

  1. Load tracked trajectories and source video.
  2. Review flagged suspicious segments in the suspicion queue.
  3. Correct identity assignments using the interactive canvas.
  4. Export refined trajectories.