arXiv subjects

Yudian Zheng

Publications and source records attributed to Yudian Zheng.

5 recordsLinked to original sources

arXiv2025

EmoCAST: Emotional Talking Portrait via Emotive Text Description

Emotional talking head synthesis aims to generate talking portrait videos with vivid expressions. Existing methods still exhibit limitations in control flexibility, motion naturalness, and expression quality. Moreover, currently available datasets are mainly collected in lab settings, further exacerbating these shortcomings and hindering real-world deployment. To address these challenges, we propose EmoCAST, a diffusion-based talking head framework for precise, text-driven emotional synthesis. Its contributions are threefold: (1) architectural modules that enable effective text control; (2) an emotional talking-head dataset that expands the framework's ability; and (3) training strategies that further improve performance. Specifically, for appearance modeling, emotional prompts are integrated through a text-guided emotive attention module, enhancing spatial knowledge to improve emotion understanding. To strengthen audio-emotion alignment, we introduce an emotive audio attention module to capture the interplay between controlled emotion and driving audio, generating emotion-aware features to guide precise facial motion synthesis. Additionally, we construct a large-scale, in-the-wild emotional talking head dataset with emotive text descriptions to optimize the framework's performance. Based on this dataset, we propose an emotion-aware sampling strategy and a progressive functional training strategy that improve the model's ability to capture nuanced expressive features and achieve accurate lip-sync. Overall, EmoCAST achieves state-of-the-art performance in generating realistic, emotionally expressive, and audio-synchronized talking-head videos. Project Page: https://github.com/GVCLab/EmoCAST

cs.CV↗

arXiv2023

Sketch Video Synthesis

Understanding semantic intricacies and high-level concepts is essential in image sketch generation, and this challenge becomes even more formidable when applied to the domain of videos. To address this, we propose a novel optimization-based framework for sketching videos represented by the frame-wise Bézier curve. In detail, we first propose a cross-frame stroke initialization approach to warm up the location and the width of each curve. Then, we optimize the locations of these curves by utilizing a semantic loss based on CLIP features and a newly designed consistency loss using the self-decomposed 2D atlas network. Built upon these design elements, the resulting sketch video showcases impressive visual abstraction and temporal coherence. Furthermore, by transforming a video into SVG lines through the sketching process, our method unlocks applications in sketch-based video editing and video doodling, enabled through video composition, as exemplified in the teaser.

cs.CV↗

arXiv2019

Towards Extracting Highlights From Recorded Live Videos: An Implicit Crowdsourcing Approach

Live streaming platforms need to store a lot of recorded live videos on a daily basis. An important problem is how to automatically extract highlights (i.e., attractive short video clips) from these massive, long recorded live videos. One approach is to directly apply a highlight extraction algorithm to video content. However, algorithmic approaches are either domain-specific, which require experts to spend a long time to design, or resource-intensive, which require a lot of training data and/or computing resources. In this paper, we propose Lightor, a novel implicit crowdsourcing approach to overcome these limitations. The key insight is to collect users' natural interactions with a live streaming platform, and then leverage them to detect highlights. Lightor consists of two major components. Highlight Initializer collects time-stamped chat messages from a live video and then uses them to predict approximate highlight positions. Highlight Extractor keeps track of how users interact with these approximate highlight positions and then refines these positions iteratively. We find that the collected user chat and interaction data are very noisy, and propose effective techniques to deal with noise. Lightor can be easily deployed into existing live streaming platforms, or be implemented as a web browser extension. We recruit hundreds of users from Amazon Mechanical Turk, and evaluate the performance of Lightor using two popular games in Twitch. The results show that Lightor can achieve high extraction precision with a small set of training data and low computing resources.

cs.DB↗

arXiv2018

Crowd-Powered Data Mining

Many data mining tasks cannot be completely addressed by auto- mated processes, such as sentiment analysis and image classification. Crowdsourcing is an effective way to harness the human cognitive ability to process these machine-hard tasks. Thanks to public crowdsourcing platforms, e.g., Amazon Mechanical Turk and Crowd- Flower, we can easily involve hundreds of thousands of ordinary workers (i.e., the crowd) to address these machine-hard tasks. In this tutorial, we will survey and synthesize a wide spectrum of existing studies on crowd-powered data mining. We first give an overview of crowdsourcing, and then summarize the fundamental techniques, including quality control, cost control, and latency control, which must be considered in crowdsourced data mining. Next we review crowd-powered data mining operations, including classification, clustering, pattern mining, machine learning using the crowd (including deep learning, transfer learning and semi-supervised learning) and knowledge discovery. Finally, we provide the emerging challenges in crowdsourced data mining.

cs.DB↗

arXiv2017

T-Crowd: Effective Crowdsourcing for Tabular Data

Crowdsourcing employs human workers to solve computer-hard problems, such as data cleaning, entity resolution, and sentiment analysis. When crowdsourcing tabular data, e.g., the attribute values of an entity set, a worker's answers on the different attributes (e.g., the nationality and age of a celebrity star) are often treated independently. This assumption is not always true and can lead to suboptimal crowdsourcing performance. In this paper, we present the T-Crowd system, which takes into consideration the intricate relationships among tasks, in order to converge faster to their true values. Particularly, T-Crowd integrates each worker's answers on different attributes to effectively learn his/her trustworthiness and the true data values. The attribute relationship information is also used to guide task allocation to workers. Finally, T-Crowd seamlessly supports categorical and continuous attributes, which are the two main datatypes found in typical databases. Our extensive experiments on real and synthetic datasets show that T-Crowd outperforms state-of-the-art methods in terms of truth inference and reducing the cost of crowdsourcing.

cs.DB↗