Projects

Face Recognition
	This project contains such capabilities as face detection, attribute analysis, face comparison, and liveness detection. It is flexibly used for financial, PAN security, and retail industry scenarios. It meets business requirements, such as identity verification, human face attendance, and gate access. Face detection: It can detect the human face rapidly and return to the human face box position. Attribute analysis: it can identify several kinds of attribute information accurately, including but not limited to age, gender, mask. Liveness Detection: it offers six on-line and off-line liveness detection abilities. Thus, it can judge whether the operation is done by a real human and prevent cheating behaviours, such as using a picture, video, or mock-up.

Face Recognition

This project contains such capabilities as face detection, attribute analysis, face comparison, and liveness detection. It is flexibly used for financial, PAN security, and retail industry scenarios. It meets business requirements, such as identity verification, human face attendance, and gate access.

Face detection: It can detect the human face rapidly and return to the human face box position.

Attribute analysis: it can identify several kinds of attribute information accurately, including but not limited to age, gender, mask.

Liveness Detection: it offers six on-line and off-line liveness detection abilities. Thus, it can judge whether the operation is done by a real human and prevent cheating behaviours, such as using a picture, video, or mock-up.

Secure Federated Learning Framework
	This project aims to unify and implement state-of-the-art adversarial attacks and defenses under federated learning scenarios. Adversarial attacks aim to breach privacy through various kinds of approaches, including but not limited to poisoning data or model, modifying loss function, model inversion, etc. Specifically, in this project, we cover the following attack strategies: 1. Gradient Attacks 2. Backdoor Attacks 3. Byzantine Attacks In contrast, defense strategies aim to build a secure and robust FL framework, which enables resilience to various kinds of failures and attacks. In this project, we cover the following defense strategies: 1. Differential Privacy 2. Gradient Compression 3. Robust Aggregation Code (unavailable now)

Secure Federated Learning Framework

This project aims to unify and implement state-of-the-art adversarial attacks and defenses under federated learning scenarios.

Adversarial attacks aim to breach privacy through various kinds of approaches, including but not limited to poisoning data or model, modifying loss function, model inversion, etc. Specifically, in this project, we cover the following attack strategies:
1. Gradient Attacks
2. Backdoor Attacks
3. Byzantine Attacks

In contrast, defense strategies aim to build a secure and robust FL framework, which enables resilience to various kinds of failures and attacks. In this project, we cover the following defense strategies:
1. Differential Privacy
2. Gradient Compression
3. Robust Aggregation

Code (unavailable now)

Awesome Federated Machine Learning
	Everything about federated learning, the goal is to keep tracking the latest research advancements of federated learning, including but not limited to research papers, books, codes, tutorials, and videos. 1. Top Machine Learning Conferences (ICML, ICLR, NeurIPS) 2. Top Computer Vision Conferences (CVPR, ICCV, ECCV) 3. Top Artificial Intelligence and Data Mining Conferences (AAAI, AISTATS, KDD) 4. Books 5. Papers with different research directions 6. Conference workshops 7. Videos, Tutorials, Blogs, and Open sources Code: Github Project

Awesome Federated Machine Learning

Everything about federated learning, the goal is to keep tracking the latest research advancements of federated learning, including but not limited to research papers, books, codes, tutorials, and videos.

1. Top Machine Learning Conferences (ICML, ICLR, NeurIPS)
2. Top Computer Vision Conferences (CVPR, ICCV, ECCV)
3. Top Artificial Intelligence and Data Mining Conferences (AAAI, AISTATS, KDD)
4. Books
5. Papers with different research directions
6. Conference workshops
7. Videos, Tutorials, Blogs, and Open sources

Code: Github Project

Anbu Huang

Projects