Learn new objects fast with few examples without forget old ones by hierarchical learning with HDA

Improve training of object segmentation with only boxes by using mask quality with BoxTeacher

Get object pose with self-supervised learning on videos with self-pose

Better rare object recognition by balancing gradients with EFL

Real-time dynamic 3D object segmentation including non-rigid objects without priors with STAR

Faster and better multi-person pose estimation by using a single stage with AdaptivePose

Get 3D scene in real-time from monocular images using depth uncertainty with Rosinol

Improve monocular depth estimation without labeled training data by image masking with MIMDepth

Segment objects in new domain by unsupervised training with boxes in old domain with WUDA

Improve object detection under difficult conditions by learning many image pre-processors with GDIP

Better enhancement of dim noisy images by using subspaces with RLED-Net

Detect object changes between images despite different viewpoint and lighting with CYWS

Explain an image classifier result using neural-symbolic learning with Greybox XAI

Survey of super-resolution advances including diffusion and transformers

Survey of deep learning for high-resolution computer vision

Survey of crowd counting with computer vision

Video correspondence by self-supervised learning with less cost by using spatial then time with Li

Get background even when have a moving camera using transformers with DeepMCBM

Simpler unsupervised object segmentation in videos using spectral clustering with SSL-VOS

Get optical flow and object orientation even with lighting changes with INV-Flow2PoseNet