Results 161 to 170 of about 374,244 (266)
This paper presents a degeneracy‐aware light detection and ranging (LiDAR)‐inertial framework that enhances LiDAR simultaneous localization and mapping performance in challenging environments. The proposed system integrates a dual‐layer robust odometry frontend with a Scan‐Context‐based loop‐closure detection backend.
Haoming Yang +4 more
wiley +1 more source
Representation learning approach for understanding structured documents. [PDF]
Trivedi A +3 more
europepmc +1 more source
Musculoskeletal humanoids exhibit rich biomechanical properties that remain insufficiently unified in prior discussions. This article systematically categorizes muscle characteristics into five properties: redundancy, independency, anisotropy, variable moment arm, and nonlinear elasticity, and analyzes their combined effects on control.
Kento Kawaharazuka +2 more
wiley +1 more source
LDA-DETR: A lightweight dynamic attention-enhanced DETR for small object detection. [PDF]
Shi Y, Li J, Jia Y, Hong Q.
europepmc +1 more source
FTGRN introduces an LLM‐enhanced framework for gene regulatory network inference through a two‐stage workflow. It combines a Transformer‐based model, pretrained on GPT‐4 derived gene embeddings and regulatory knowledge, with a fine‐tuning stage utilizing single‐cell RNA‐seq data.
Guangzheng Weng +7 more
wiley +1 more source
Lightweight traffic sign detection algorithm with noise suppression and semantic enhancement. [PDF]
Wang L, Ai Q, Jin H.
europepmc +1 more source
Adaptive multi‐indicator contrastive predictive coding is introduced as a self‐supervised pretraining framework for multivariate EHR time series. An adaptive sliding‐window algorithm and 2D convolutional neural network encoder capture localized temporal patterns and global indicator dependencies, enabling label‐efficient disease prediction that ...
Hongxu Yuan +3 more
wiley +1 more source
FDSNet: dynamic multimodal fusion stage selection for autonomous driving via feature disagreement scoring. [PDF]
Mohammed A, Ibrahim HM, Omar NM.
europepmc +1 more source
Robust Dysarthric Speech Recognition with GAN Enhancement and LLM Correction
This study tackles dysarthric speech recognition by combining generative adversarial network (GAN)‐generated synthetic data with large language model (LLM)‐based error correction. The approach integrates three key elements: an improved CycleGAN to generate synthetic dysarthric speech for data augmentation, a multimodal automatic speech recognition core
Yibo He +3 more
wiley +1 more source
Autonomous navigation in unstructured outdoor environments using semantic segmentation guided reinforcement learning. [PDF]
Tibermacine A +4 more
europepmc +1 more source

