2 | USTC-AC

TED: Towards Explainable and Holistically Aligned Empathetic Dialogue Generation

Sat, 27 Jun 2026 17:53:00 +0800

Empathetic Response Generation Through Multi-modality

Thu, 14 Aug 2025 17:53:00 +0800

Facial Action Unit Recognition Guided by Labeling Rules

Tue, 11 Mar 2025 17:53:00 +0800

From Traits to Empathy: Personality-Aware Multimodal Empathetic Response Generation

Sat, 18 Jan 2025 17:53:00 +0800

Integrating Visual Modalities with Large Language Models for Mental Health Support

Sat, 18 Jan 2025 17:53:00 +0800

Facial Action Unit Recognition Enhanced by Text Descriptions of FACS

Mon, 30 Sep 2024 17:53:00 +0800

VAD: A Video Affective Dataset with Danmu

Thu, 28 Mar 2024 09:53:00 +0000

A Multi-Stage Visual Perception Approach for Image Emotion Analysis

Wed, 27 Mar 2024 09:50:00 +0000

Pose-robust personalized facial expression recognition through unsupervised multi-source domain adaptation

Mon, 19 Feb 2024 09:50:00 +0000

Pose-Aware Facial Expression Recognition Assisted by Expression Descriptions

Sun, 16 Jul 2023 09:53:00 +0000

Emotional Attention Detection and Correlation Exploration for Image Emotion Distribution Learning

Tue, 06 Apr 2021 11:54:51 +0000

Capturing Emotion Distribution for Multimedia Emotion Tagging

Thu, 18 Feb 2021 11:58:00 +0000

Multimedia collections usually induce multiple emotions in audiences. The data distribution of multiple emotions can be leveraged to facilitate the learning process of emotion tagging, yet has not been thoroughly explored. To address this, we propose adversarial learning to fully capture emotion distributions for emotion tagging of multimedia data. The proposed multimedia emotion tagging approach includes an emotion classifier and a discriminator. The emotion classifier predicts emotion labels of multimedia data from their content. The discriminator distinguishes the predicted emotion labels from the ground truth labels. The emotion classifier and the discriminator are trained simultaneously in competition with each other. By jointly minimizing the traditional supervised loss and maximizing the distribution similarity between the predicted emotion labels and the ground truth emotion labels, the proposed multimedia emotion tagging approach successfully captures both the mapping function between multimedia content and emotion labels as well as prior distribution in emotion labels, and thus achieves state-of-the-art performance for multiple emotion tagging, as demonstrated by the experimental results on four benchmark databases.

Fig. The framework of the proposed multi-emotion tagging model. C is the multiple emotion tag classifier, and D is the multiple emotion tag discriminator. C(x) is the multiple emotion prediction for feature vector x. y is the ground truth multiple emotion label of x. y 0 is the real multiple emotion label from ground truth label set. The dotted line indicates that there is an one-to-one correspondence between C(x) and y. See text for details.

Exploring Adversarial Learning for Deep Semi-Supervised Facial Action Unit Recognition

Fri, 01 Jan 2021 00:00:00 +0000

Multi-task face analyses through adversarial learning

Fri, 01 Jan 2021 00:00:00 +0000

Dual Learning for Facial Action Unit Detection Under Nonfull Annotation

Mon, 13 Jul 2020 09:53:00 +0000

A Novel Dynamic Model Capturing Spatial and Temporal Patterns for Facial Expression Analysis

Wed, 01 Jan 2020 00:00:00 +0000

Capturing Joint Label Distribution for Multi-Label Classification Through Adversarial Learning

Wed, 01 Jan 2020 00:00:00 +0000

Exploring Domain Knowledge for Facial Expression-Assisted Action Unit Activation Recognition

Wed, 01 Jan 2020 00:00:00 +0000

Knowledge-Augmented Multimodal Deep Regression Bayesian Networks for Emotion Video Tagging

Wed, 01 Jan 2020 00:00:00 +0000

Pose-aware Adversarial Domain Adaptation for Personalized Facial Expression Recognition

Wed, 01 Jan 2020 00:00:00 +0000

Posed and Spontaneous Expression Distinction Using Latent Regression Bayesian Networks

Wed, 01 Jan 2020 00:00:00 +0000

Affective Computing for Large-Scale Heterogeneous Multimedia Data: A Survey

Tue, 01 Jan 2019 00:00:00 +0000

Capturing Feature and Label Relations Simultaneously for Multiple Facial Action Unit Recognition

Tue, 01 Jan 2019 00:00:00 +0000

Content-Based Video Emotion Tagging Augmented by Users' Multiple Physiological Responses

Tue, 01 Jan 2019 00:00:00 +0000

Facial Action Unit Recognition and Intensity Estimation Enhanced Through Label Dependencies

Tue, 01 Jan 2019 00:00:00 +0000

Multiple Face Analyses through Adversarial Learning

Tue, 01 Jan 2019 00:00:00 +0000

Weakly Supervised Dual Learning for Facial Action Unit Recognition

Tue, 01 Jan 2019 00:00:00 +0000

Learning with privileged information for multi-Label classification

Mon, 01 Jan 2018 00:00:00 +0000

Thermal Augmented Expression Recognition

Mon, 01 Jan 2018 00:00:00 +0000

Weakly Supervised Facial Action Unit Recognition With Domain Knowledge

Mon, 01 Jan 2018 00:00:00 +0000

Expression-assisted facial action unit recognition under incomplete AU annotation

Sun, 01 Jan 2017 00:00:00 +0000

Feature and label relation modeling for multiple-facial action unit classification and intensity estimation

Sun, 01 Jan 2017 00:00:00 +0000

Learning with Privileged Information for Multi-Label Classification

Sun, 01 Jan 2017 00:00:00 +0000

Capturing global spatial patterns for distinguishing posed and spontaneous expressions

Fri, 01 Jan 2016 00:00:00 +0000

Facial expression recognition through modeling age-related spatial patterns

Fri, 01 Jan 2016 00:00:00 +0000

Gender recognition from visible and thermal infrared facial images

Fri, 01 Jan 2016 00:00:00 +0000

Emotion Recognition with the Help of Privileged Information