UTOPY trains unrolling algorithms for ill-posed inverse problems via a fidelity homotopy path from synthetic well-posed to real ill-posed sensing operators, yielding up to 2.5 dB PSNR gains.
U-net: Convolutional networks for biomedical image segmentation,
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
representative citing papers
Dino U-Net combines a frozen DINOv3 backbone with an adapter and fidelity-aware projection module to achieve state-of-the-art medical image segmentation across seven public datasets.
TrackNet, a heatmap-based CNN trained on single or multiple frames, tracks tennis balls in videos achieving 99.7% precision on one match and 95.3% in 10-fold cross-validation.
FPCNet uses an encoder-decoder architecture with MD and SEU modules to learn multi-context crack features and achieves faster, more accurate pixel-level detection than prior methods on CFD and G45 datasets.
Presents and compares U-Net and DenseU-Net models for fully automatic tongue-contour segmentation in ultrasound images, reporting comparable accuracy with differences in speed and cross-dataset generalization.
citing papers explorer
-
UTOPY: Unrolling Algorithm Learning via Fidelity Homotopy for Inverse Problems
UTOPY trains unrolling algorithms for ill-posed inverse problems via a fidelity homotopy path from synthetic well-posed to real ill-posed sensing operators, yielding up to 2.5 dB PSNR gains.
-
Dino U-Net: Exploiting High-Fidelity Dense Features from Foundation Models for Medical Image Segmentation
Dino U-Net combines a frozen DINOv3 backbone with an adapter and fidelity-aware projection module to achieve state-of-the-art medical image segmentation across seven public datasets.
-
TrackNet: A Deep Learning Network for Tracking High-speed and Tiny Objects in Sports Applications
TrackNet, a heatmap-based CNN trained on single or multiple frames, tracks tennis balls in videos achieving 99.7% precision on one match and 95.3% in 10-fold cross-validation.
-
FPCNet: Fast Pavement Crack Detection Network Based on Encoder-Decoder Architecture
FPCNet uses an encoder-decoder architecture with MD and SEU modules to learn multi-context crack features and achieves faster, more accurate pixel-level detection than prior methods on CFD and G45 datasets.
-
A CNN-based tool for automatic tongue contour tracking in ultrasound images
Presents and compares U-Net and DenseU-Net models for fully automatic tongue-contour segmentation in ultrasound images, reporting comparable accuracy with differences in speed and cross-dataset generalization.