U-Net
PulseAugur coverage of U-Net — every cluster mentioning U-Net across labs, papers, and developer communities, ranked by signal.
- used by Sentinel-2 70%
- used by DagsHub 70%
- instance of ScienceCast 70%
- used by Diffusion Transformer 70%
- used by diffusion model 70%
- instance of alphaXiv 70%
- instance of Gotit.pub 70%
- uses Diffusion Transformer 70%
- competes with Deeplabv3 Plus 60%
- used by Grad-CAM++ 60%
- competes with ResNet-50 60%
- instance of multilayer perceptron 60%
18 day(s) with sentiment data
-
New self-supervised method enhances CT scan reconstruction without ground-truth data
Researchers have developed a self-supervised method called Noise2Inverse Learned Primal-Dual (N2I-LPD) to improve X-ray computed tomography reconstruction. This new approach allows for the training of reconstruction ope…
-
CALHippo uses ML to map 3D brain cell structures
Researchers have developed CALHippo, a novel system for mapping neurons and glial cells in the human brain's hippocampus in 3D. The system utilizes state-of-the-art segmentation networks, like CellPoseSAM, to identify a…
-
New diffusion model enhances skin lesion segmentation accuracy
Researchers have developed MLFFM-SegDiff, a novel diffusion model designed to improve the segmentation of skin lesions in dermoscopic images. This model addresses challenges such as blurred boundaries and artifacts by i…
-
New CCUA method boosts AI image generation for rare classes
Researchers have developed a new method called Contrastive Conditional-Unconditional Alignment (CCUA) to improve the quality and diversity of images generated by diffusion models, particularly for classes with limited t…
-
New grid-size-invariant neural networks offer faster rock-fluid interaction modeling
Researchers have developed eight new surrogate models to predict fluid flow in porous media, aiming to reduce the computational expense of traditional high-fidelity numerical models. Four of these are reduced-order mode…
-
FPGA accelerator boosts U-Net energy efficiency with merged arithmetic · 3 sources tracked
Researchers have developed an energy-efficient hardware accelerator for U-Net's convolutional layers, implemented on a field-programmable gate array (FPGA). The proposed merged multiply-add (MMA) architecture fuses oper…
-
Blasto-Net: AI model for blastocyst analysis in IVF · 2 sources tracked
Researchers have developed Blasto-Net, a novel multi-task deep learning model designed for comprehensive blastocyst analysis in in vitro fertilization (IVF). This model simultaneously performs segmentation of key compar…
-
Deep Learning Framework Aids Cervical Cancer Detection in Pap Smear Analysis
Researchers have developed a deep learning framework to aid in the early detection of cervical cancer using Pap smear images. The system integrates U-Net for image segmentation and a classification model, tested on the …
-
New research explores audio-visual and flow-matching techniques for speech enhancement
Two new research papers explore advanced techniques for speech enhancement using generative models. The first paper introduces Audio-visual Contrastive Alignment (AVCA) to improve diffusion-based speech enhancement by e…
-
New framework PeLAP-A prunes latent diffusion models, revealing 'sparsity collapse'
Researchers have introduced PeLAP-A, a framework designed to make latent diffusion models more lightweight by adaptively pruning unimportant channels in the latent space. This method uses a multilayer perceptron to pred…
-
New AI method improves breast ultrasound lesion segmentation specificity
Researchers have developed a new method for segmenting lesions in breast ultrasound images, addressing challenges like boundary leakage and false-positive activations. The approach uses entropy-guided boundary supervisi…
-
Native-space AI models outperform template-based for subcortical brain segmentation
Researchers have developed two U-Net-based pipelines for segmenting subcortical brain regions, specifically the Subthalamic Nucleus (STN), Red Nucleus (RN), and Substantia Nigra (SN), which are critical for neurosurgica…
-
AlphaEarth and TESSERA embeddings show promise for fine-scale climate zone mapping
A new study published on arXiv explores the use of AlphaEarth and TESSERA embeddings for fine-scale Local Climate Zone (LCZ) mapping in Switzerland. Researchers compared these embeddings with traditional Sentinel-1/2 co…
-
Deep neural networks enhance urban temperature forecasting with high-resolution satellite data
Researchers have developed deep neural network models to improve the resolution and forecasting of urban land surface temperatures. By combining data from geostationary and polar-orbiting satellites, they created models…
-
MC Dropout's reliability in brain tumor segmentation questioned
Researchers have investigated the reliability of Monte Carlo Dropout (MC Dropout) for segmenting brain tumors in MRI scans, finding that while it can align uncertainty with errors, it may not always guarantee clinical s…
-
New framework enhances infrared small target detection using SCR prior
Researchers have developed a new framework called REEM (Reweighted Explicit-visibility Enhanced Modulation) to improve infrared small target detection. This method incorporates Signal-to-Clutter Ratio (SCR) as a prior d…
-
New method enhances medical image segmentation for skin lesions
Researchers have developed PEFT-MedSAM, a parameter-efficient fine-tuning method for the Medical Segment Anything Model (MedSAM) to improve the segmentation of skin lesions in dermoscopic images. This technique freezes …
-
New SEGS method tackles Janus problem in text-to-3D generation
Researchers have developed a new framework called Structural Energy-Guided Sampling (SEGS) to address the Janus problem in text-to-3D generation. This issue causes inconsistent geometry across different viewpoints. SEGS…
-
Spacecraft Perception Model Achieves Top Ranking in SPARK 2026 Challenge
Researchers have developed a novel segmentation-based detection method for multi-task spacecraft perception, addressing challenges like limited annotated data and difficult visual conditions. Their compact architecture,…
-
New AI models offer faster, more efficient medical image translation
Researchers have developed new methods for medical image translation that are faster and more efficient than existing diffusion models. One study introduces a lightweight U-Net that outperforms a state-of-the-art Denois…