# cjm-byte-track
<!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->
A standalone Python implementation of the
[ByteTrack](https://arxiv.org/abs/2110.06864) multi-object tracker based
on the [official
implementation](https://github.com/ifzhang/ByteTrack/tree/main/yolox/tracker).
## Install
``` sh
pip install cjm_byte_track
```
## Tutorial:
- [**Real-Time Object Tracking with YOLOX and
ByteTrack:**](https://christianjmills.com/posts/pytorch-train-object-detector-yolox-tutorial/byte-track/)
Learn how to track objects across video frames with YOLOX and
ByteTrack.
## How to use
``` python
# Import ByteTrack package
from cjm_byte_track.core import BYTETracker
from cjm_byte_track.matching import match_detections_with_tracks
```
``` python
# Initialize a ByteTracker object
tracker = BYTETracker(track_thresh=0.25, track_buffer=30, match_thresh=0.8, frame_rate=frame_fps)
with tqdm(total=frames, desc="Processing frames") as pbar:
while video_capture.isOpened():
ret, frame = video_capture.read()
if ret:
# Prepare an input image for inference
rgb_img, input_dims, offsets, min_img_scale, input_img = prepare_image_for_inference(frame, test_sz, max_stride)
# Convert the existing input image to NumPy format
input_tensor_np = np.array(input_img, dtype=np.float32).transpose((2, 0, 1))[None]/255
# Start performance counter`m
start_time = time.perf_counter()
# Run inference
outputs = session.run(None, {"input": input_tensor_np})[0]
# Process the model output
proposals = process_outputs(outputs, input_tensor_np.shape[input_dim_slice], bbox_conf_thresh)
# Apply non-max suppression to the proposals with the specified threshold
proposal_indices = nms_sorted_boxes(calc_iou(proposals[:, :-2]), iou_thresh)
proposals = proposals[proposal_indices]
bbox_list = (proposals[:,:4]+[*offsets, 0, 0])*min_img_scale
label_list = [class_names[int(idx)] for idx in proposals[:,4]]
probs_list = proposals[:,5]
# Update tracker with detections.
track_ids = [-1]*len(bbox_list)
# Convert to tlbr format
tlbr_boxes = bbox_list.copy()
tlbr_boxes[:, 2:4] += tlbr_boxes[:, :2]
# Update tracker with detections
tracks = tracker.update(
output_results=np.concatenate([tlbr_boxes, probs_list[:, np.newaxis]], axis=1),
img_info=rgb_img.size,
img_size=rgb_img.size)
track_ids = match_detections_with_tracks(tlbr_boxes=tlbr_boxes, track_ids=track_ids, tracks=tracks)
# End performance counter
end_time = time.perf_counter()
# Calculate the combined FPS for object detection and tracking
fps = 1 / (end_time - start_time)
# Display the frame rate in the progress bar
pbar.set_postfix(fps=fps)
# Filter object detections based on tracking results
bbox_list, label_list, probs_list, track_ids = zip(*[(bbox, label, prob, track_id)
for bbox, label, prob, track_id
in zip(bbox_list, label_list, probs_list, track_ids) if track_id != -1])
# Annotate the current frame with bounding boxes and tracking IDs
annotated_img = draw_bboxes_pil(
image=rgb_img,
boxes=bbox_list,
labels=[f"{track_id}-{label}" for track_id, label in zip(track_ids, label_list)],
probs=probs_list,
colors=[int_colors[class_names.index(i)] for i in label_list],
font=font_file,
)
annotated_frame = cv2.cvtColor(np.array(annotated_img), cv2.COLOR_RGB2BGR)
video_writer.write(annotated_frame)
pbar.update(1)
else:
break
video_capture.release()
video_writer.release()
```
Raw data
{
"_id": null,
"home_page": "https://github.com/cj-mills/cjm-byte-track",
"name": "cjm-byte-track",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": "",
"keywords": "nbdev jupyter notebook python",
"author": "Christian Mills",
"author_email": "millscj@protonmail.com",
"download_url": "https://files.pythonhosted.org/packages/5a/1e/8e3cab0086bbeb1f07e24ee965f5e0a84091c05b7b8f9d05f6327351d1b2/cjm-byte-track-0.0.6.tar.gz",
"platform": null,
"description": "# cjm-byte-track\n\n<!-- WARNING: THIS FILE WAS AUTOGENERATED! DO NOT EDIT! -->\n\nA standalone Python implementation of the\n[ByteTrack](https://arxiv.org/abs/2110.06864) multi-object tracker based\non the [official\nimplementation](https://github.com/ifzhang/ByteTrack/tree/main/yolox/tracker).\n\n## Install\n\n``` sh\npip install cjm_byte_track\n```\n\n## Tutorial:\n\n- [**Real-Time Object Tracking with YOLOX and\n ByteTrack:**](https://christianjmills.com/posts/pytorch-train-object-detector-yolox-tutorial/byte-track/)\n Learn how to track objects across video frames with YOLOX and\n ByteTrack.\n\n## How to use\n\n``` python\n# Import ByteTrack package\nfrom cjm_byte_track.core import BYTETracker\nfrom cjm_byte_track.matching import match_detections_with_tracks\n```\n\n``` python\n# Initialize a ByteTracker object\ntracker = BYTETracker(track_thresh=0.25, track_buffer=30, match_thresh=0.8, frame_rate=frame_fps)\n\nwith tqdm(total=frames, desc=\"Processing frames\") as pbar:\n while video_capture.isOpened():\n ret, frame = video_capture.read()\n if ret:\n \n # Prepare an input image for inference\n rgb_img, input_dims, offsets, min_img_scale, input_img = prepare_image_for_inference(frame, test_sz, max_stride)\n \n # Convert the existing input image to NumPy format\n input_tensor_np = np.array(input_img, dtype=np.float32).transpose((2, 0, 1))[None]/255\n\n # Start performance counter`m\n start_time = time.perf_counter()\n \n # Run inference\n outputs = session.run(None, {\"input\": input_tensor_np})[0]\n\n # Process the model output\n proposals = process_outputs(outputs, input_tensor_np.shape[input_dim_slice], bbox_conf_thresh)\n \n # Apply non-max suppression to the proposals with the specified threshold\n proposal_indices = nms_sorted_boxes(calc_iou(proposals[:, :-2]), iou_thresh)\n proposals = proposals[proposal_indices]\n \n bbox_list = (proposals[:,:4]+[*offsets, 0, 0])*min_img_scale\n label_list = [class_names[int(idx)] for idx in proposals[:,4]]\n probs_list = proposals[:,5]\n\n # Update tracker with detections.\n track_ids = [-1]*len(bbox_list)\n\n # Convert to tlbr format\n tlbr_boxes = bbox_list.copy()\n tlbr_boxes[:, 2:4] += tlbr_boxes[:, :2]\n\n # Update tracker with detections\n tracks = tracker.update(\n output_results=np.concatenate([tlbr_boxes, probs_list[:, np.newaxis]], axis=1),\n img_info=rgb_img.size,\n img_size=rgb_img.size)\n track_ids = match_detections_with_tracks(tlbr_boxes=tlbr_boxes, track_ids=track_ids, tracks=tracks)\n\n # End performance counter\n end_time = time.perf_counter()\n # Calculate the combined FPS for object detection and tracking\n fps = 1 / (end_time - start_time)\n # Display the frame rate in the progress bar\n pbar.set_postfix(fps=fps)\n\n # Filter object detections based on tracking results\n bbox_list, label_list, probs_list, track_ids = zip(*[(bbox, label, prob, track_id) \n for bbox, label, prob, track_id \n in zip(bbox_list, label_list, probs_list, track_ids) if track_id != -1])\n\n # Annotate the current frame with bounding boxes and tracking IDs\n annotated_img = draw_bboxes_pil(\n image=rgb_img, \n boxes=bbox_list, \n labels=[f\"{track_id}-{label}\" for track_id, label in zip(track_ids, label_list)],\n probs=probs_list,\n colors=[int_colors[class_names.index(i)] for i in label_list], \n font=font_file,\n )\n annotated_frame = cv2.cvtColor(np.array(annotated_img), cv2.COLOR_RGB2BGR)\n \n video_writer.write(annotated_frame)\n pbar.update(1)\n else:\n break\nvideo_capture.release()\nvideo_writer.release()\n```\n",
"bugtrack_url": null,
"license": "Apache Software License 2.0",
"summary": "A standalone Python implementation of the ByteTrack multi-object tracker based on the official implementation.",
"version": "0.0.6",
"project_urls": {
"Homepage": "https://github.com/cj-mills/cjm-byte-track"
},
"split_keywords": [
"nbdev",
"jupyter",
"notebook",
"python"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "b3b96e251ea133163a517f6d32044fcbd99e4cb423e7c62b02b2118ac6fa4b23",
"md5": "71d5c4b2f6a01a1fc66c20bffe86674c",
"sha256": "045dcc0f91fe423a065cbd1d85eff1217868b09ac9a32ee01a0f89d56eb1e501"
},
"downloads": -1,
"filename": "cjm_byte_track-0.0.6-py3-none-any.whl",
"has_sig": false,
"md5_digest": "71d5c4b2f6a01a1fc66c20bffe86674c",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 20378,
"upload_time": "2023-10-27T19:42:54",
"upload_time_iso_8601": "2023-10-27T19:42:54.876651Z",
"url": "https://files.pythonhosted.org/packages/b3/b9/6e251ea133163a517f6d32044fcbd99e4cb423e7c62b02b2118ac6fa4b23/cjm_byte_track-0.0.6-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "5a1e8e3cab0086bbeb1f07e24ee965f5e0a84091c05b7b8f9d05f6327351d1b2",
"md5": "d818573b112f20fe159a3b1222fda924",
"sha256": "98f9f876014e3bd245966b71addde1abbec95f51e33e18f0fea6b380e13d0ddb"
},
"downloads": -1,
"filename": "cjm-byte-track-0.0.6.tar.gz",
"has_sig": false,
"md5_digest": "d818573b112f20fe159a3b1222fda924",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 18077,
"upload_time": "2023-10-27T19:42:56",
"upload_time_iso_8601": "2023-10-27T19:42:56.436870Z",
"url": "https://files.pythonhosted.org/packages/5a/1e/8e3cab0086bbeb1f07e24ee965f5e0a84091c05b7b8f9d05f6327351d1b2/cjm-byte-track-0.0.6.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-10-27 19:42:56",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "cj-mills",
"github_project": "cjm-byte-track",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "cjm-byte-track"
}
JSON
