draw bboxes on a video

annotate.py

Here is a Python program that uses OpenCV to generate a video with bounding boxes and lines for the driven trajectories on top of the original video footage:
```
import cv2
import numpy as np

# Load the MOT challenge data file
mot_data = []
with open('mot_data.txt', 'r') as f:
    for line in f:
        frame, id, x, y, w, h, _, _, _, _ = line.strip().split(',')
        mot_data.append({
            'frame': int(frame),
            'id': int(id),
            'x': int(float(x)),
            'y': int(float(y)),
            'w': int(float(w)),
            'h': int(float(h))
        })

# Load the video capture
cap = cv2.VideoCapture('input_video.mp4')

# Get the video properties
fps = cap.get(cv2.CAP_PROP_FPS)
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))

# Create a video writer
fourcc = cv2.VideoWriter_fourcc(*'XVID')
out = cv2.VideoWriter('output_video.avi', fourcc, fps, (width, height))

# Create a dictionary to store the trajectories
trajectories = {}

# Iterate through the frames
while True:
    ret, frame = cap.read()
    if not ret:
        break

    # Get the current frame number
    frame_num = int(cap.get(cv2.CAP_PROP_POS_FRAMES))

    # Draw bounding boxes and trajectories
    for detection in mot_data:
        if detection['frame'] == frame_num:
            x, y, w, h = detection['x'], detection['y'], detection['w'], detection['h']
            cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 0), 2)

            # Add the detection to the trajectory
            if detection['id'] not in trajectories:
                trajectories[detection['id']] = []
            trajectories[detection['id']].append((x, y))

    # Draw trajectories
    for trajectory in trajectories.values():
        for i in range(1, len(trajectory)):
            cv2.line(frame, trajectory[i-1], trajectory[i], (0, 0, 255), 2)

    # Write the output frame
    out.write(frame)

# Release the video capture and writer
cap.release()
out.release()
```
This program assumes that the MOT challenge data file is in the same format as the MOT16 dataset, with each line representing a detection in the format `frame,id,x,y,w,h,...`. The program loads the video capture, reads the MOT challenge data file, and iterates through the frames. For each frame, it draws bounding boxes around the detected objects and connects the detections to form trajectories. The output video is written to a file named `output_video.avi`.

Note that this program assumes that the video and MOT challenge data file are in the same directory as the Python script. You may need to modify the file paths and names to match your specific use case. Additionally, you can customize the appearance of the bounding boxes and trajectories by modifying the `cv2.rectangle` and `cv2.line` function calls.