Evaluate

Use eval to score tracking runs on MOT-style datasets with TrackEval-backed metrics.

Examples

!!! example

=== "CLI"

    ```bash
    boxmot eval --benchmark mot17 --split ablation --tracker boosttrack --verbose
    ```

=== "Python"

    ```python
    from boxmot import Boxmot

    boxmot = Boxmot(detector="yolov8n", reid="lmbn_n_duke", tracker="boosttrack")
    metrics = boxmot.val(benchmark="mot17", split="ablation")
    print(metrics)
    ```

Typical workflow

!!! example

=== "CLI"

    For repeated experiments:

    ```bash
    boxmot generate --benchmark mot17 --split ablation
    boxmot eval --benchmark mot17 --split ablation --tracker boosttrack
    ```

    This lets `eval` reuse precomputed detections and embeddings.

=== "Python"

    ```python
    from boxmot import Boxmot

    boxmot = Boxmot(detector="yolov8n", reid="lmbn_n_duke", tracker="boosttrack")
    metrics = boxmot.val(benchmark="mot17", split="ablation")
    print(metrics)
    ```

Public detections

Use --detection-source to run with public MOTChallenge detections instead of the benchmark's configured detector:

boxmot eval --benchmark mot17 --split ablation --tracker boosttrack --detection-source frcnn
boxmot eval --benchmark mot17 --split ablation --tracker boosttrack --detection-source sdp
boxmot eval --benchmark mot17 --split ablation --tracker boosttrack --detection-source dpm

--detection-source public uses the default public detector defined in the benchmark YAML. When omitted (or --detection-source private), eval runs the configured detector model.

See Benchmark Workflows for details on how public detections are resolved.

Kalman filter noise tuning

Use --tune-kf to estimate per-sequence Kalman filter process and measurement noise (Q/R matrices) from the cached detections and ground truth before tracking:

boxmot eval --benchmark mot17 --split ablation --tracker boosttrack --tune-kf

This is most useful for trackers with Kalman-filter-based motion models. It requires cached detections and ground truth to be available.

For runtime adaptation without ground truth, use --adaptive-kf instead, which estimates noise online via the Mehra (1970) method.

Postprocessing

!!! example

=== "CLI"

    `eval` can apply optional postprocessing before scoring.
    Multiple steps can be chained with commas and are applied sequentially to the same result files:

    ```bash
    # Single step
    boxmot eval --benchmark mot17 --split ablation --tracker boosttrack --postprocessing gsi

    # Chained: GSI runs first, then GTA reads GSI's output
    boxmot eval --benchmark mot17 --split ablation --tracker boosttrack --postprocessing gsi,gta
    ```

    Available steps:

    | Step | Description |
    | --- | --- |
    | `gsi` | Gaussian-smoothed interpolation — fills gaps and smooths trajectories |
    | `gbrc` | Gradient-boosting reconnection — ML-based interpolation and smoothing |
    | `gta` | Global tracklet association — offline split-and-connect across the full sequence |

=== "Python"

    `Boxmot.val(...)` is the Python-facing validation entry point. Postprocessing details and metric interpretation are the same as in the CLI evaluation pipeline.

!!! warning "Chained steps overwrite in place" When chaining multiple postprocessing steps, each step reads the MOT result files, transforms them, and writes back to the same directory. The second step operates on the output of the first.

See Evaluation and Postprocessing.

See Benchmark Workflows for cache reuse, MMOT benchmark ids, and replay image-loading behavior.

Native C++ replay

Use --tracker-backend cpp to run the cached replay stage through a native tracker implementation:

boxmot eval --benchmark mot17 --split ablation --tracker bytetrack --tracker-backend cpp
boxmot eval --benchmark mot17 --split ablation --tracker ocsort:cpp

Native replay is currently available for botsort, bytetrack, ocsort, occluboost, and sfsort. --tracking-backend cpp is still accepted as a compatibility alias, but --tracker-backend cpp is the canonical selector.

Main outputs

• combined benchmark metrics such as HOTA, MOTA, and IDF1
• per-sequence summaries
• optional runtime timing summary with --show-timing
• MOT-style tracker outputs
• reused cache paths and evaluation artifacts in the run directory

See Evaluation and Postprocessing.

CLI Arguments

::: mkdocs-click :module: boxmot.engine.cli :command: boxmot :depth: 1 :command: eval :style: table :prog_name: boxmot eval