#deep learning
120 resultsDeep learning brain conductivity mapping using a patch-based 3D U-net
Nils Hampe, Ulrich Katscher, Cornelis A. T. van den Berg +2
The paper presents a patch‑based 3D U‑Net that learns to predict brain conductivity maps from B1 transceive phase MRI data, evaluating performance on simulated and in‑vivo datasets…
X-Section: Cross-Section Prediction for Enhanced RGBD Fusion
Andrea Nicastro, Ronald Clark, Stefan Leutenegger
The paper introduces X-Section, a deep‑learning method that predicts per‑object thickness from RGB‑D images and integrates these predictions into an extended KinectFusion pipeline…
Towards Deep Learning-Based EEG Electrode Detection Using Automatically Generated Labels
Nils Gessert, Martin Gromniak, Marcel Bengs +2
The paper proposes a deep‑learning system that detects EEG electrodes in RGB‑D images, using a robotic setup to automatically generate large amounts of training labels.
Explicit Shape Encoding for Real-Time Instance Segmentation
Wenqiang Xu, Haiyang Wang, Fubo Qi +1
The paper introduces ESE‑Seg, a top‑down instance segmentation framework that encodes object shapes explicitly using an inner‑center radius signature and Chebyshev polynomial fitti…
GenerationMania: Learning to Semantically Choreograph
Zhiyu Lin, Kyle Xiao, Mark Riedl
The paper proposes a deep neural network approach that automatically creates rhythm-game charts for arbitrary music by predicting which sounds the player should hit and assigning c…
Enhanced 3D convolutional networks for crowd counting
Zhikang Zou, Huiliang Shao, Xiaoye Qu +2
The paper introduces a temporal channel-aware (TCA) block that uses 3D convolutions and channel-wise modulation to capture spatio‑temporal information for more accurate crowd count…
3-D Positioning and Environment Mapping for mmWave Communication Systems
Jie Yang, Shi Jin, Chao-Kai Wen +2
The paper develops a 3‑D joint positioning, velocity estimation, and environment mapping (SLAM) framework for millimeter‑wave cloud radio access networks, using hybrid delay‑angle…
Deep Non-Rigid Structure from Motion
Chen Kong, Simon Lucey
The paper presents a deep neural network that jointly estimates camera poses and 3D shape from 2D point tracks for non‑rigid objects, handling large numbers of images and complex d…
Unsupervised Neural Quantization for Compressed-Domain Similarity Search
Stanislav Morozov, Artem Babenko
The paper proposes an unsupervised deep neural network that compresses visual descriptors using multi-codebook quantization for fast image retrieval.
Emotion Dependent Facial Animation from Affective Speech
Rizwan Sadiq, Sasan AsadiAbadi, Engin Erzin
The paper proposes a two‑stage deep learning system that first classifies speech into seven emotions and then generates corresponding facial shape animations, showing improved perf…
Efficient Structurally-Strengthened Generative Adversarial Network for MRI Reconstruction
Wenzhong Zhou, Huiqian Du, Wenbo Mei +1
The paper introduces ESSGAN, a generative adversarial network with structurally strengthened connections and residual-in-residual blocks, designed to reconstruct high‑quality MRI i…
Bayesian Loss for Crowd Count Estimation with Point Supervision
Zhiheng Ma, Xing Wei, Xiaopeng Hong +1
The paper introduces a Bayesian loss function that uses point annotations to directly supervise crowd count estimation, improving accuracy over traditional density‑map approaches w…
Recent Advances in Deep Learning for Object Detection
Xiongwei Wu, Doyen Sahoo, Steven C. H. Hoi
The paper surveys recent deep‑learning approaches for visual object detection, reviewing detector architectures, learning strategies, and benchmark applications.
Processamento de linguagem natural em Português e aprendizagem profunda para o domÃnio de Ãleo e Gás
Diogo Gomes, Alexandre Evsukoff
The paper reviews deep learning techniques for natural language processing in Portuguese, focusing on challenges and applications within the oil and gas domain.
FAKEDETECTOR: Effective Fake News Detection with Deep Diffusive Neural Network
Jiawei Zhang, Bowen Dong, Philip S. Yu
FAKEDETECTOR is a model that detects fake news by jointly learning representations of news articles, creators, and subjects using a deep diffusive neural network built from explici…
BERT-based Ranking for Biomedical Entity Normalization
Zongcheng Ji, Qiang Wei, Hua Xu
The paper proposes a biomedical entity normalization system that fine‑tunes pre‑trained BERT variants (BERT, BioBERT, ClinicalBERT) and shows that these models achieve higher accur…
A Fast and Precise Method for Large-Scale Land-Use Mapping Based on Deep Learning
Xuan Yang, Zhengchao Chen, Baipeng Li +3
The paper presents a fast, deep‑learning based approach using an optimized convolutional neural network to produce high‑resolution land‑use maps over large areas, achieving 81.5% a…
No Need of Data Pre-processing: A General Framework for Radio-Based Device-Free Context Awareness
Bo Wei, Kai Li, Chengwen Luo +2
The paper proposes a deep‑learning framework that directly processes raw radio signals for device‑free context awareness, removing the need for separate preprocessing or feature ex…
Distinguishing Individual Red Pandas from Their Faces
Qi He, Qijun Zhao, Ning Liu +3
The paper presents a deep‑learning framework for automatically recognizing individual red pandas from facial photographs, and evaluates it on a newly created red panda image databa…
Deep Learning based Wearable Assistive System for Visually Impaired People
Yimin Lin, Kai Wang, Wanxin Yi +1
The paper presents a wearable assistive system that uses deep learning on RGB‑D data to generate safe navigation instructions and 3D semantic maps for visually impaired users.
Hyper Vision Net: Kidney Tumor Segmentation Using Coordinate Convolutional Layer and Attention Unit
D. Sabarinathan, M. Parisa Beham, S. M. Md. Mansoor Roomi
The paper introduces Hyper Vision Net, a deep learning model that uses coordinate convolution and attention mechanisms to improve kidney tumor segmentation in CT scans.
Bayesian Inference for Large Scale Image Classification
Jonathan Heek, Nal Kalchbrenner
The paper introduces ATMC, an adaptive noise MCMC algorithm for Bayesian inference in deep neural networks, and demonstrates its ability to improve classification accuracy and unce…
ABD-Net: Attentive but Diverse Person Re-Identification
Tianlong Chen, Shaojin Ding, Jingyi Xie +5
The paper introduces ABD-Net, a person re-identification model that combines complementary attention modules with an orthogonality-based diversity regularization to learn more disc…
Human Perceptual Evaluations for Image Compression
Yash Patel, Srikar Appalaraju, R. Manmatha
The paper conducts human user studies to show that deep‑learning image compression methods optimized for higher MS‑SSIM scores can actually look worse to people than traditional co…