sense
hagrid
sense | hagrid | |
---|---|---|
4 | 7 | |
726 | 487 | |
0.0% | - | |
0.0 | 5.3 | |
over 2 years ago | about 1 month ago | |
Python | Python | |
MIT License | - |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
sense
-
Can not Download 20BN Jester Dateset
Because of the fact that Qualcomm acquired 20 billion neurons (20bn), jester dataset can't be downloaded from their website. Does anyone know an alternative source to download?
- Sense: A New Open-Source Video Understanding Framework for Deep Learning
-
[P] Sense: Open Source Framework for Video Understanding & Action Recognition with Deep Learning
For counting, the Sense project provides a tool for temporally annotating your videos. This way you can tag key positions in the movement (like up and down position for pushups), which the model will also learn to output and can then easily be counted. Here's an example script for counting jumping jacks and squats.
hagrid
-
COMPLETE OVERRIDE, THE WORKER. Reality on the second play. The keyframes were created at full size directly in the txt2img tab of Stable Diffusion all at the same time. It took about 30 minutes.
Not sure how HaGRID is going but one would expected hands problem would be already solved by now.
-
Magda s Jitkou a jejich mech před Ústavem robotiky v Praze někdy na konci srpna roku 1982 ;-) #midjourneyart
Pokud sháníš detailní ruce, tak HaGRID by mohl pomoct
-
Meirl
For hands and fingers, if you know how to inpaint and/or integrate datasets into your own model, HaGRID is by far the best dataset and there are a few models you can use for basic inpainting/correction. For other limbs you need just a better model and/or learn negative prompts better. The Stable Diffusion sub can help with both.
-
Are there any known hand-trained models?
This may be useful? https://github.com/hukenovs/hagrid
-
GREAT NEWS!: Automagic 1111 added support for Stable Diffusion 2.0 - thank you so much !!!
By the way, this is the hands-model Emad mentioned yesterday they want to start training on for SD: https://github.com/hukenovs/hagrid
-
HaGRID (HAnd Gesture Recognition Image Dataset)
Github
- HaGRID is a large image dataset for hand gesture recognition systems
What are some alternatives?
transfer-learning-conv-ai - 🦄 State-of-the-Art Conversational AI with Transfer Learning
gluon-cv - Gluon CV Toolkit
SSD-pytorch - SSD: Single Shot MultiBox Detector pytorch implementation focusing on simplicity
pytorch-grad-cam - Advanced AI Explainability for computer vision. Support for CNNs, Vision Transformers, Classification, Object detection, Segmentation, Image similarity and more.
NeuralCDE - Code for "Neural Controlled Differential Equations for Irregular Time Series" (Neurips 2020 Spotlight)
chitra - A multi-functional library for full-stack Deep Learning. Simplifies Model Building, API development, and Model Deployment.
pytorch-forecasting - Time series forecasting with PyTorch
label-studio - Label Studio is a multi-type data labeling and annotation tool with standardized output format
pytorch-tutorial - PyTorch Tutorial for Deep Learning Researchers
autogluon - Fast and Accurate ML in 3 Lines of Code
bbox-visualizer - Make drawing and labeling bounding boxes easy as cake