---

Latest Works

An Embarrassingly Simple Approach for Intellectual Property Rights Protection on Recurrent Neural Networks Star

This paper proposes a practical approach for the IPR protection on recurrent neural networks (RNN) with the Gatekeeper concept.

DeepIPR: Deep Neural Network Ownership Verification with Passports Star

We propose novel passport-based DNN ownership verification schemes which are both robust to network modifications and resilient to ambiguity attacks (i.e. the DNN model performance of an original task will be significantly deteriorated due to forged passports). Extension of NeurIPS 2019 (acceptance rate: 1428/6743 ~ 21.18%).

CyEDA: Cycle-object Edge Consistency Domain Adaptation Star

This paper proposed CyEDA to perform global level domain adaptation (DA) without any pre-trained networks integration/annotation labels.

ProX: A Reversed Once-for-All Network Training Paradigm for Efficient Edge Models Training in Medical Imaging

We propose a reversed OFA Network training algorithm - Progressive Expansion (ProX) that achieve up to 68% training time reduction.

Extremely Low-light Image Enhancement with Scene Text Restoration

This work deals with the problem of low-light image enhancement, in the context of scene texts.

One Loss for All: Deep Hashing with a Single Cosine Similarity based Learning Objective Star

This paper proposes a novel deep hashing model with only a single learning objective which is a simplification from most state of the art papers generally use lots of losses and regularizer.

Protecting Intellectual Property of Generative Adversarial Networks from Ambiguity Attacks Star

This paper presents a complete protection framework in both black-box and white-box settings to enforce IPR protection on GANs.