Tao Li

Purdue University | Purdue ·  Department of Computer Science

With billions of personal images being generated from social media and cameras of all sorts on a daily basis, security and privacy are unprecedentedly challenged. Although extensive attempts have been made, existing face image de-identification techniques are either insufficient in photo-reality or incapable of balancing privacy and usability quali...

The advent of connected and autonomous vehicles (CAVs) will change driving behavior and the travel environment, providing opportunities for safer, smoother, and smarter road transportation. During the transition from the current human-driven vehicles (HDVs) to a fully CAV traffic environment, the road traffic will consist of a “mixed” traffic flow...

The concept of beauty has been debated by philosophists and psychologists for centuries, but most definitions are subjective and metaphysical, and deficit in accuracy, generality, and scalability. In this paper, we present a novel study on mining beauty semantics of facial attributes based on big data, with an attempt to objectively construct descr...

Vehicle-to-vehicle communications can be unreliable as interference causes communication failures. Thereby, the information flow topology for a platoon of Connected Autonomous Vehicles (CAVs) can vary dynamically. This limits existing Cooperative Adaptive Cruise Control (CACC) strategies as most of them assume a fixed information flow topology (IFT...

With one billion monthly viewers, and millions of users discussing and sharing opinions, comments below YouTube videos are rich sources of data for opinion mining and sentiment analysis. We introduce the YouTube AV 50K dataset, a freely-available collections of more than 50,000 YouTube comments and metadata below autonomous vehicle (AV)-related vid...

There is a growing privacy concern due to the popularity of social media and surveillance systems, along with advances in face recognition software. However, established image obfuscation techniques are either vulnerable to re-identification attacks by human or deep learning models, insufficient in preserving image fidelity, or too computationally...

There is growing concern about image privacy due to the popularity of social media and photo devices, along with increasing use of face recognition systems. However, established image de-identification techniques are either too subject to re-identification, produce photos that are insufficiently realistic, or both. To tackle this, we present a nove...

To make the best use of the underlying structure of faces, the collective information through face datasets and the intermediate estimates during the upsampling process, here we introduce a fully convolutional multi-stage neural network for 4$\times$ super-resolution for face images. We implicitly impose facial component-wise attention maps using a...

Most existing models for information flow propagation in a vehicle-to-vehicle (V2V) communications environment are descriptive. They lack capabilities to control information flow, which may preclude their ability to meet application needs, including the need to propagate different information types simultaneously to different target locations withi...

Large volumes of medical data, such as Electronic Medical Records (EMRs), provide promising sources for knowledge discovery. However, existing literature mostly focuses on building a knowledge graph based on correlations and association rules, which merely output Markovian equivalent classes and fail to provide insights of the causal mechanisms beh...

In an effort to enhance privacy while balance usability and reality of facial images, we propose a novel framework called AnonymousNet, which encompasses four stages: (i) facial attribute prediction by leveraging a deep Convolutional Neural Network (CNN); (ii) facial semantic obfuscation directed by privacy metrics; (iii) natural image synthesis us...

This work showcases a new approach for causal discovery by leveraging user experiments and recent advances in photo-realistic image editing, demonstrating a potential of identifying causal factors and understanding complex systems counterfactually. We introduce the beauty learning problem as an example, which has been discussed metaphysically for c...

Emergent cooperative adaptive cruise control (CACC) strategies being proposed for platoon formation in the connected autonomous vehicle (CAV) context mostly assume idealized fixed information flow topologies (IFTs) for the platoon, implying guaranteed vehicle-to-vehicle (V2V) communications for the IFT assumed. In reality, V2V communications are un...

The advent of connected autonomous vehicles provides opportunities for safer, smoother, and smarter transportation. However, broadcasting information to surrounding vehicles and infrastructures risks security and privacy. Moreover, control decisions relying on such information are vulnerable to malicious attacks. In this paper , we propose a cooper...

Most existing models for information flow propagation in a vehicle-to-vehicle (V2V) communications environment are descriptive. They lack capabilities to control information flow, which may preclude their ability to meet application needs, including the need to propagate different information types simultaneously to different target locations withi...

Emergent cooperative adaptive cruise control (CACC) strategies being proposed for platoon formation in the connected autonomous vehicle (CAV) context mostly assume idealized fixed information flow topologies (IFTs) for the platoon, implying guaranteed vehicle-to-vehicle (V2V) communications for the IFT assumed. In reality, V2V communications are un...

Enormous online textual information provide intriguing opportunities for under-standings of social and economic semantics. In this paper, we propose a novel text regression model based on a conditional generative adversarial network (GAN), with an attempt to associate textual data and social outcomes in an semi-supervised manner. Besides promising...

Enormous online textual information provides intriguing opportunities for understandings of social and economic semantics. In this paper, we propose a novel text regression model based on a conditional generative adversarial network (GAN), with an attempt to associate textual data and social outcomes in a semi-supervised manner. Besides promising p...

DSRC-based V2X communication network has higher risk of failures, including device errors and hijacking. In the worst case, the V2X network may undergo Byzantine failures. We propose a novel V2X communication system that applies the first provably Byzantine-tolerant aggregation rule for vehicular ad hoc networks (VANET). The system guarantees relia...

Recommendation systems that automatically generate personalized music playlists for users have attracted tremendous attention in recent years. Nowadays, most music recommendation systems rely on item-based or user-based collaborative filtering, or content-based approaches. In this paper, we propose a novel mixture hidden Markov model (HMM) for musi...

Recommendation systems that automatically generate personalized music playlists for users have attracted tremendous attention in recent years. Nowadays, most music recommendation systems rely on item-based or user-based collaborative filtering or content-based approaches. In this paper, we propose a novel mixture hidden Markov model (HMM) for music...

Generative adversarial networks (GANs) promote recent successes of deep learning in fields such as computer vision and speech synthesis. However, training a GAN is notoriously tricky and unpredictable, and requires substantial efforts from both human and machines. In this paper, we introduce a novel Bayesian framework based on recent advances in de...

The advent of connected and autonomous vehicles (CAVs) will change driving behavior and the travel environment, providing opportunities for safer, smoother, and smarter road transportation. During the transition from the current human-driven vehicles (HDVs) to a fully CAV traffic environment, the road traffic will consist of a "mixed" traffic flow...

Emergent cooperative adaptive cruise control (CACC) strategies being proposed in the literature for platoon formation in the Connected Autonomous Vehicle (CAV) context mostly assume idealized fixed information flow topologies (IFTs) for the platoon, implying guaranteed vehicle-to-vehicle (V2V) communications for the IFT assumed. Since CACC strategi...

Motivations of this talk include: (i) briefly share fancy stuff in CVPR 2018 and AVS 2018; (ii) discuss current trends in both academia and industry; and (iii) potential research directions and collaboration opportunities.

A brief version has been uploaded on Research Gate. Full version can be found on: https://taoli.me/publications/li2018high.pdf

A survey of single image super-resolution (SISR) techniques from a modern viewpoint.

YoutubeStat.py is a Python module to download YouTube statistics and rankings from www.kedoo.com. More information can be found in the project homepage: https://github.com/Eroica-cpp/YouTube-Statistics/

The discipline of data science has become increasingly popular in these days. The Web, as one of the most important data sources, offers tremendous potential, and web crawling techniques are intriguing and promising. In this tutorial, I compare multiple self-learning resources for Python beginners in Step One. In Step Two, a real-world coding examp...

A Fortran library for K-means algorithm written in Fortran 90. Parallel computing techniques are applied to deal with high-dimensional sparse data with a focus of real-world Natural Language Processing (NLP) problems.