SEER (self-supervised vision model from Facebook)
SEER is a self-supervised vision model developed by Facebook that enhances image recognition without the need for labeled data.
computer vision
CrowdPose
CrowdPose is a dataset and method for estimating human poses in crowded scenes, enhancing the performance of pose estimation algorithms.
MPII Human Pose (pose estimation dataset)
The MPII Human Pose dataset is a widely used benchmark for evaluating human pose estimation algorithms, featuring diverse images and detailed annotations.
ViT (vision transformer)
ViT (Vision Transformer) is a deep learning architecture that applies the transformer model, originally designed for natural language processing, to computer vision tasks. It processes images by dividing them into patches and treating these patches as tokens, enabling the use of self-attention mechanisms for image understanding.
DINOv2
DINOv2 is a self-supervised learning model designed for visual representation learning, building on its predecessor DINO for improved performance.
Mixture of depths
Mixture of depths refers to the combination or layering of different depth levels within a medium or context, often used in fields such as image processing, geology, and data visualization to represent or analyze complex structures or scenes.
DECA (detailed expression capture and animation)
DECA (detailed expression capture and animation) is a technology and framework used in computer graphics for capturing and animating highly detailed facial expressions. It enables realistic and high-fidelity facial animations by reconstructing 3D facial geometry and expressions from images or video.
Human mesh recovery (HMR)
Human Mesh Recovery (HMR) is a technique in computer vision for reconstructing 3D human body meshes from images or video data.
Computer vision
Computer vision is a multidisciplinary field that enables computers to interpret and process visual information from the world. It involves the development of algorithms and systems that can analyze images and videos to extract meaningful data.
SMPL (skinned multi-person linear model)
The Skinned Multi-Person Linear (SMPL) model is a widely used parametric human body model that captures the shape and pose of individuals in a 3D space.