Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

task/4041-change-yolov5-to-yolov8 #151

Merged
merged 2 commits into from
Jan 25, 2024
Merged

task/4041-change-yolov5-to-yolov8 #151

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4578c49
replace yolov5 with yolov8 where usefull
frunika Jan 25, 2024
23992ad
Change yolov8s.pt to lower case
martinbaerwolff Jan 25, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions docs/OTLabels/index.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,12 +3,12 @@
> The **backbone** of OpenTrafficCam.

[OTLabels](https://github.com/OpenTrafficCam/OTLabels) is a workflow for retraining AI models to detect road users in videos based on custom video frames
and existing open source software packages ([CVAT](https://github.com/openvinotoolkit/cvat), [YOLO v5](https://github.com/ultralytics/yolov5)).
and existing open source software packages ([CVAT](https://github.com/openvinotoolkit/cvat), [YOLO v8](https://github.com/ultralytics/ultralytics)).

## Key features

* Annotation of custom video frames using existing open source software ([CVAT](https://github.com/openvinotoolkit/cvat))
* Retrain existing neural network models with [YOLO v5](https://github.com/ultralytics/yolov5) for individual optimization of the detection ability
* Retrain existing neural network models with [YOLO v8](https://github.com/ultralytics/ultralytics) for individual optimization of the detection ability
* Runs on any Windows laptop. However, a decent GPU with CUDA support is strongly recommended.

## Content of documentation
Expand Down
2 changes: 1 addition & 1 deletion docs/OTVision/index.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,7 @@ graph LR
(using [ffmpeg](https://ffmpeg.org/))
* [Detection](usage/detect) (joint localization and classification) of road users using
state-of-the-art AI object detection models in single video frames
(currently using [YOLOv5 by Ultralytics](https://github.com/ultralytics/yolov5))
(currently using [YOLOv8 by Ultralytics](https://github.com/ultralytics/ultralytics))
* [Tracking](usage/track) of detected road users over multiple frames
(currently using the
[IOU-Tracker by Bochinski et al.](https://github.com/bochinski/iou-tracker))
Expand Down
10 changes: 5 additions & 5 deletions docs/OTVision/usage/detect.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -86,13 +86,13 @@ or

Name of weights from PyTorch hub or path to weights file.

This parameter is optional and defaults to `YOLOv5s.pt`.
This parameter is optional and defaults to `yolov8s.pt`.

### conf

`--conf <float>`

The YOLOv5 model confidence threshold.
The YOLOv8 model confidence threshold.
Should be a float value between zero and one.

The confidence threshold is the minimum confidence for a detection to be considered
Expand All @@ -104,7 +104,7 @@ This parameter is optional and defaults to `0.25`.

`--iou <float>`

The YOLOv5 model IOU threshold.
The YOLOv8 model IOU threshold.
Should be a float value between zero and one.

The IOU threshold is the overlap threshold for areas of bounding boxes used in
Expand All @@ -116,7 +116,7 @@ This parameter is optional and defaults to `0.45`.

`--chunksize <int>`

The number of frames of a video to be detected by YOLOv5 in one iteration.
The number of frames of a video to be detected by YOLOv8 in one iteration.
Should be an integer above zero.

This parameter is optional and defaults to `1`.
Expand All @@ -133,7 +133,7 @@ This parameter is optional and defaults to `--no-half`.

### force

`--force` to force a reload of a YOLOv5 standard model from PyTorch hub instead of
`--force` to force a reload of a YOLOv8 standard model from PyTorch hub instead of
using a cached model from previous detection runs.

`--no-force` to prevent forcing this reload.
Expand Down