Picture for Shi Yin

Shi Yin

Serialized Output Training by Learned Dominance

Add code
Jul 04, 2024
Viaarxiv icon

A Comprehensive Investigation on Speaker Augmentation for Speaker Recognition

Add code
Jun 11, 2024
Viaarxiv icon

A Framework of SO-equivariant Non-linear Representation Learning and its Application to Electronic-Structure Hamiltonian Prediction

Add code
May 10, 2024
Viaarxiv icon

Harmonizing Covariance and Expressiveness for Deep Hamiltonian Regression in Crystalline Material Research: a Hybrid Cascaded Regression Framework

Add code
Jan 15, 2024
Viaarxiv icon

1DFormer: Learning 1D Landmark Representations via Transformer for Facial Landmark Tracking

Add code
Nov 01, 2023
Viaarxiv icon

Spot keywords from very noisy and mixed speech

Add code
May 28, 2023
Viaarxiv icon

Attentive One-Dimensional Heatmap Regression for Facial Landmark Detection and Tracking

Add code
Apr 11, 2020
Figure 1 for Attentive One-Dimensional Heatmap Regression for Facial Landmark Detection and Tracking
Figure 2 for Attentive One-Dimensional Heatmap Regression for Facial Landmark Detection and Tracking
Figure 3 for Attentive One-Dimensional Heatmap Regression for Facial Landmark Detection and Tracking
Figure 4 for Attentive One-Dimensional Heatmap Regression for Facial Landmark Detection and Tracking
Viaarxiv icon

Multiple Face Analyses through Adversarial Learning

Add code
Nov 18, 2019
Figure 1 for Multiple Face Analyses through Adversarial Learning
Figure 2 for Multiple Face Analyses through Adversarial Learning
Figure 3 for Multiple Face Analyses through Adversarial Learning
Figure 4 for Multiple Face Analyses through Adversarial Learning
Viaarxiv icon

Closed-Loop Adaptation for Weakly-Supervised Semantic Segmentation

Add code
May 29, 2019
Figure 1 for Closed-Loop Adaptation for Weakly-Supervised Semantic Segmentation
Figure 2 for Closed-Loop Adaptation for Weakly-Supervised Semantic Segmentation
Figure 3 for Closed-Loop Adaptation for Weakly-Supervised Semantic Segmentation
Figure 4 for Closed-Loop Adaptation for Weakly-Supervised Semantic Segmentation
Viaarxiv icon

Fast and accurate reconstruction of HARDI using a 1D encoder-decoder convolutional network

Add code
Mar 21, 2019
Figure 1 for Fast and accurate reconstruction of HARDI using a 1D encoder-decoder convolutional network
Figure 2 for Fast and accurate reconstruction of HARDI using a 1D encoder-decoder convolutional network
Figure 3 for Fast and accurate reconstruction of HARDI using a 1D encoder-decoder convolutional network
Viaarxiv icon