MobileViT: an accurate, light-weight, mobile-friendly vision transformer MobileViT: Light-weight, General-purpose, and Mobile-friendly...

Detect 133 points on face, hands, and feet in real-time by using neighborhood Keypoint Communities arXiv paper abstract...

Fast online action detection by storing past data relevant now Information Elevation Network for Fast Online Action Detection arXiv paper...

Better interactive image segmentation for training by using edge information EdgeFlow: Achieving Practical Interactive Segmentation with...

Flexible artifact removal in JPEG images Towards Flexible Blind JPEG Artifacts Removal arXiv paper abstract https://arxiv.org/abs/2109.14...

Non-uniform image sampling for better image segmentation Learning to Downsample for Segmentation of Ultra-High Resolution Images arXiv...

Better deblurring of motion images by using edges Edge Prior Augmented Networks for Motion Deblurring on Naturally Blurry Images arXiv...

Survey of advances in continual learning in computer vision Recent Advances of Continual Learning in Computer Vision: An Overview arXiv...

Replace people and objects in street scene images including proper shadows Repopulating Street Scenes arXiv paper abstract...

From image directly output text labels and coordinates of detected objects Pix2seq: A Language Modeling Framework for Object Detection...

Reinforcement learning for learning multi-step tasks on new objects in images Example-Driven Model-Based Reinforcement Learning for...

3DETR transformer for 3D Object Detection An End-to-End Transformer Model for 3D Object Detection arXiv paper abstract...

Real-time face distance and iris track on mobile phone without depth sensor MediaPipe Iris: Real-time Iris Tracking & Depth Estimation...

Better 3D pose estimates in video by dynamically learning joint relationships Learning Dynamical Human-Joint Affinity for 3D Pose...

Unsupervised learning of image classes from dynamic video stream Online Unsupervised Learning of Visual Representations and Categories...

Real-time 3D hand reconstruction from a single monocular image Towards Accurate Alignment in Real-time 3D Hand-Mesh Reconstruction arXiv...

Get depth, regions, and layout from panoramic image quickly and accurately with horizontal features HoHoNet: 360 Indoor Holistic...

Image classification without normalization that is faster and better than with normalization High-Performance Large-Scale Image...

Image segmentation of objects and regions using transformers Panoptic SegFormer arXiv paper abstract https://arxiv.org/abs/2109.03814...

Using an audio and vision transformer to count crowds Audio-Visual Transformer Based Crowd Counting arXiv paper abstract...