Vitpose with yolov5 and tracking
Project description
ViTPose-Pytorch no mmcv needed
ViTPose (https://github.com/ViTAE-Transformer/ViTPose) + ByteTrack (https://github.com/ifzhang/ByteTrack) + yolov5 (https://github.com/ultralytics/yolov5)
Preparation
Download the official weights from ViTPose. Create a dictionary called models. Tested with both the vitpose-b models.
mkdir models
mv <weight_path> ~/ViTPose-Pytorch/models/
Requirements
If nvidia-tensorrt is not found install nvidia-pyindex first.
pip3 insltall -r requirements.txt
How to run
Export ViTPose TRT model
python3 export.py --include engine --device 0
Export yolov5 engine, either by cloning the official yolov5 repo and using the export function or via cd /.cache/torch/hub/yolov5_master and exporting from there. Move the engine file into the root folder of ViTPose-Pytorch.
With a webcam or another device
ls /dev/vid*
3 demo examples are provided . To use this lib as a package run
python3 -m build
Install the whl in the dist folder via pip.
Examples
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file vitpose_infer-0.2.0.tar.gz.
File metadata
- Download URL: vitpose_infer-0.2.0.tar.gz
- Upload date:
- Size: 100.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.16
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
d3567c6a1b534ec8c1bb0ea3f04f3617141fccbd2c9fdfd583f65931b1f90269
|
|
| MD5 |
bb5a0cd31072baf5456cbd40272a396a
|
|
| BLAKE2b-256 |
e3ccb5b5047b32779bb18db2f767b95de8aaedfa169cde6697520e440d3bcd06
|
File details
Details for the file vitpose_infer-0.2.0-py3-none-any.whl.
File metadata
- Download URL: vitpose_infer-0.2.0-py3-none-any.whl
- Upload date:
- Size: 155.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.2 CPython/3.9.16
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
9f9515e664c429f8aab7aca7821838e0caca70adcf1f69d507cbda290cd0c1fe
|
|
| MD5 |
cec96f6bfadfce5e9ce00dbb616e2633
|
|
| BLAKE2b-256 |
3f1feac0af8d21af7485adadbf474fdc99cd3bd9166c0623bafdad80783c54dc
|
