UPerNet
PulseAugur coverage of UPerNet — every cluster mentioning UPerNet across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New Contrastive Uncertainty Learning improves MRI segmentation accuracy
Researchers have developed a novel framework called Contrastive Uncertainty Learning (CUL) to improve the accuracy of anatomical structure segmentation in prostate MRIs. CUL utilizes predictive uncertainty to guide repr…
-
New research compares AI vision models to human physical reasoning
Researchers have developed a new method to compare the object representations learned by computer vision models with those of humans. Using a time-to-collision and change detection task with 178 and 50 human participant…
-
New HAFR-Net framework improves VHR remote sensing image segmentation
Researchers have developed a new framework called HAFR-Net for segmenting very-high-resolution remote sensing images. This network adaptively organizes and refines hierarchical representations from pretrained encoders, …
-
Vision foundation model adapted for high-resolution flood mapping
Researchers have developed a method for flood mapping using a vision foundation model adapted from satellite imagery. The model, named Prithvi-2.0-UPN, was fine-tuned on RGB datasets and demonstrated state-of-the-art re…
-
Vision foundation model advances flood mapping from RGB imagery
Researchers have developed a new method for flood mapping using airborne RGB imagery, leveraging a vision foundation model called Prithvi-2.0-UPN. This model, pre-trained on satellite data, demonstrates strong generaliz…
-
AI advances medical image segmentation with new frameworks and techniques · 8 sources tracked
Researchers are developing advanced AI frameworks for medical image segmentation, focusing on improving accuracy and efficiency. Hi-Seg enhances the Segment Anything Model (SAM) for pulmonary nodule segmentation through…
-
LALE architecture boosts land-cover estimation efficiency
Researchers have developed LALE, a novel lightweight transformer architecture designed for efficient land-cover estimation from remote sensing imagery. This architecture bifurcates its encoder, using ConvMixer stages fo…