DeepLabCut Python Colab Tensorflow Keras - Horse Racing Video Analysis
Budget: £20 – £250 GBP
Introduction
I need a video analysis ML model building in Google Colab that I can re-use to analyse my own videos.
This model needs to produce key points on videos of horses walking. Based on training data that will be provided and is pre-labelled.
Deeplabcut Model Zoo
First you should get yourself familiar with the DeepLabCut Model Zoo here - DLC Model Zoo
Use the “Horse Side View” model and analyse this video below:
Horse Video 1
Notice in the outputs that the model zoo outputs a key point video, scattergram and csv file. I need all 3 outputs from my model also.
Read here for more information on the DLC Model Zoo
Horse10 model
Then I need a model training on some pre-labelled data. This should be a combination of the model zoo above and the data contained in the Horse10 folder.
You need to download the Horse10 data here.
Please read the instructions about the Horse10 model here.
PLEASE READ LINK BELOW.
https://docs.google.com/document/d/1HqkMcMGDDNdmp2utO2My-PK0C6mtfrD1mXHjVsZgjgE/edit?usp=sharing
I need a video analysis ML model building in Google Colab that I can re-use to analyse my own videos.
This model needs to produce key points on videos of horses walking. Based on training data that will be provided and is pre-labelled.
Deeplabcut Model Zoo
First you should get yourself familiar with the DeepLabCut Model Zoo here - DLC Model Zoo
Use the “Horse Side View” model and analyse this video below:
Horse Video 1
Notice in the outputs that the model zoo outputs a key point video, scattergram and csv file. I need all 3 outputs from my model also.
Read here for more information on the DLC Model Zoo
Horse10 model
Then I need a model training on some pre-labelled data. This should be a combination of the model zoo above and the data contained in the Horse10 folder.
You need to download the Horse10 data here.
Please read the instructions about the Horse10 model here.
PLEASE READ LINK BELOW.
https://docs.google.com/document/d/1HqkMcMGDDNdmp2utO2My-PK0C6mtfrD1mXHjVsZgjgE/edit?usp=sharing