Trajectory Forecast

😍😍😍 You can use any Ultralytics-supported model here. 😍😍😍

Trajectory Forecast is a lightweight, modular extension built on top of Ultralytics YOLO that enables real-time multi-object tracking with future motion prediction. It combines detection, tracking, motion history modeling, and velocity-based forecasting into a unified pipeline that can be used both as a command-line tool and as a Python library. The system is designed for practical computer vision applications such as traffic analytics, surveillance systems, robotics pipelines, and edge AI deployments.

• Installation
• Usage
  • CLI
  • Python
• Forecasting methodology
• Project structure
• Contribute

trajectory-forecasting-demo-github.mp4

Installation

pip install trajectory-forecast

Usage

CLI

Run tracking and forecasting on a video.

trajectory-forecast --model yolo26n.pt --source "https://tinyurl.com/2f3yrppv" --output result.mp4 --show --save

If you want to adjust tracking and forecasting configuration, create a config.yaml in the directory and paste the mentioned content:

conf: 0.5                   # object detection confidence threshold
tracker: "bytetrack.yaml"   # tracker selection, i.e., "botsort.yaml" or "bytetrack.yaml"
classes: [2, 3, 5]          # classes for object detection
history: 40                 # store tracking history for number of frames
min_points: 8               # minimum tracking history to start calculating forecasting
forecast_steps: 35          # total steps for forecasting, > 40 can cause gitter effect.
vel_window: 10              # previous frames used to estimate the object's velocity.
ema_alpha: 0.7              # used to smooth the velocity or trajectory prediction.
forecast_color: [255, 0, 0] # Forecast point color (B, G, R)

After that, you can run the code using the command mentioned below.

trajectory-forecast --model yolo26n.pt --source "https://tinyurl.com/2f3yrppv" --config "path/to/config.yaml"

Python

from tf import run_inference
from tf.config import ForecastConfig

config = ForecastConfig(conf=0.5, forecast_steps=50, ema_alpha=0.7, classes=[0, 2, 5, 6, 7])

run_inference(model_path="yolo26s.pt", source="https://tinyurl.com/2f3yrppv", output_path="output.mp4", config=config)

Forecasting methodology

The current forecasting implementation is based on:

• Exponential moving average smoothing of object centers
• Median velocity estimation over a sliding window
• Linear projection of future positions
• Stationary gating to prevent unstable predictions

This approach provides a stable and computationally efficient baseline suitable for real-time systems.

Project structure

tf/
│
├── config.py        # Configuration system
├── drawing.py       # Visualization utilities
├── forecasting.py   # Velocity estimation and projection
├── tracker.py       # Track history management
├── inference.py     # Core pipeline
└── cli.py           # Command-line interface
└── utils.py         # For downloading assets from GitHub.

Contributing

The contributions are always welcome. If you would like to extend the forecasting models or improve tracking integration, please open an issue or submit a pull request.