Semantic feature extraction based on subspace learning with temporal constraints for acoustic event recognition

  title={Semantic feature extraction based on subspace learning with temporal constraints for acoustic event recognition},
  author={Qiuying Shi and Jiqing Han},
  journal={Digit. Signal Process.},

A Taxonomy of Semantic Information in Robot-Assisted Disaster Response

A taxonomy of semantic information in robot-assisted disaster response is proposed, stimulating the research that is needed to adapt, robustify, and implement state-of-the-art AI semantics methods in the challenging conditions of disasters and first responder scenarios.

Context-Aware Text Matching Algorithm for Korean Peninsula Language Knowledge Base Based on Density Clustering

Experimental results show that the proposed text matching algorithm for Korean Peninsula language knowledge base has high matching recall rate with low space requirements and can effectively match long and short texts concurrently.



Pyramidal Temporal Pooling With Discriminative Mapping for Audio Classification

PTP and DM-PTP outperform the state-of-the-art CNNs on the audio event recognition (AER) dataset, and can achieve comparable performance on the DCASE 2018 acoustic scene classification (ASC) dataset compared with other best models in the challenge.

PyTorch: An Imperative Style, High-Performance Deep Learning Library

This paper details the principles that drove the implementation of PyTorch and how they are reflected in its architecture, and explains how the careful and pragmatic implementation of the key components of its runtime enables them to work together to achieve compelling performance.

Subspace Pooling Based Temporal Features Extraction for Audio Event Recognition

Two novel pooling based methods which try to consider both the temporal dynamics and signal reconstruction ability of temporal features in the low-dimensional subspace of Audio Event Recognition.

Sub-spectrogram Segmentation for Environmental Sound Classification via Convolutional Recurrent Neural Network and Score Level Fusion

A sub-spectrogram segmentation based ESC classification framework is proposed and the proposed Convolutional Recurrent Neural Network (CRNN) and score level fusion to jointly improve the classification accuracy are adopted.

First-Person Action Recognition With Temporal Pooling and Hilbert–Huang Transform

Simulations show that the proposed approach outperforms the main state-of-the-art methods when applied to four publicly available first-person video datasets.

Discriminative Video Representation with Temporal Order for Micro-expression Recognition

This paper employs rank pooling and l2,1-norm to obtain the discriminative video representation with temporal order and extracts Local Two-Order Gradient Pattern feature of each frame to describe the subtle information in micro-expression recognition.

Fitting Segmented Regression Models by Grid Search

A grid‐search method of fitting segmented regression curves with unknown transition points is described and compared with a standard method to provide as a by‐product a way of making reliable inferences about the abscissae of the transitions.