On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,982 papers · 148 categories

Trend · papers per month

count % of corpus

5 results for “DeepLabv3+”

Uncertainty Measures and Prediction Quality Rating for the Semantic Segmentation of Nested Multi Resolution Street Scene Imagescs.CV

Method generates uncertainty measures for street scene segmentation.

problem Reliability and uncertainty measures in semantic segmentation of street scenes.

method Nested crops, neural network segmentation, post-processing, uncertainty heat maps.

result Significant improvements in classification and regression performance.

semantic segmentation street scenes nested crops uncertainty measures DeepLabv3+

2019-04-09abs ↗pdf ↗

ACNN: a Full Resolution DCNN for Medical Image Segmentationcs.LG

Deep Convolutional Neural Networks (DCNNs) are used extensively in medical image segmentation and hence 3D navigation for robot-assisted Minimally Invasive Surgeries (MISs). However, current DCNNs usually use down sampling layers for increasing the receptive field and gaining abstract semantic information. These down s…

2019-01-26abs ↗pdf ↗

Encoding the latent posterior of Bayesian Neural Networks for uncertainty quantificationcs.CV

Efficient BNNs learn latent distributions for robust uncertainty quantification.

problem Improving robustness and uncertainty of deep neural networks.

method LP-BNN uses VAEs to learn latent distributions of BNN parameters, enabling efficient ensembles.

result LP-BNN achieves competitive results in image classification, semantic segmentation, and out-of-distribution detection.

Bayesian neural networks VAE Deep ensembles Uncertainty quantification ResNet50

2020-12-04abs ↗pdf ↗

Rethinking Pre-training and Self-trainingcs.CV

Self-training outperforms pre-training on COCO object detection and segmentation datasets.

problem The effectiveness of pre-training in improving object detection and segmentation models is limited.

method Investigated self-training as an alternative method to utilize additional data.

result Self-training consistently improves model performance across various dataset sizes and data augmentation levels.

pre-training self-training object detection segmentation computer vision

2020-06-11abs ↗pdf ↗

AnoNet: Weakly Supervised Anomaly Detection in Textured Surfacescs.LG

AnoNet detects anomalies in textured surfaces with minimal training data.

problem Automating anomaly detection in textured surfaces with limited labeled data.

method AnoNet is a compact fully convolutional network that pre-seeds with an engineered filter bank to detect anomalies from few labeled images.

result AnoNet achieves state-of-the-art performance with 94% fewer parameters and 106% improvement in F1 score.

anomaly detection weakly supervised learning texture analysis compact neural networks few-shot learning

2019-11-24abs ↗pdf ↗