DiffusionEngine: Diffusion Model is Scalable Data Engine for Object Detection

no code implementations • 7 Sep 2023 • Manlin Zhang, Jie Wu, Yuxi Ren, Ming Li, Jie Qin, Xuefeng Xiao, Wei Liu, Rui Wang, Min Zheng, Andy J. Ma

This paper reveals that the recently developed Diffusion Model is a scalable data engine for object detection.

Data Augmentation object-detection +1

Adversarial Feature Augmentation for Cross-domain Few-shot Classification

1 code implementation • 23 Aug 2022 • Yanxu Hu, Andy J. Ma

Existing methods based on meta-learning predict novel-class labels for (target domain) testing tasks via meta knowledge learned from (source domain) training tasks of base classes.

Classification Cross-Domain Few-Shot

Suppressing Static Visual Cues via Normalizing Flows for Self-Supervised Video Representation Learning

1 code implementation • 7 Dec 2021 • Manlin Zhang, Jinpeng Wang, Andy J. Ma

By modelling static factors in a video as a random variable, the conditional distribution of each latent variable becomes shifted and scaled normal.

Contrastive Learning Representation Learning +1

Removing the Background by Adding the Background: Towards Background Robust Self-supervised Video Representation Learning

2 code implementations • CVPR 2021 • Jinpeng Wang, Yuting Gao, Ke Li, Yiqi Lin, Andy J. Ma, Hao Cheng, Pai Peng, Feiyue Huang, Rongrong Ji, Xing Sun

Then we force the model to pull the feature of the distracting video and the feature of the original video closer, so that the model is explicitly restricted to resist the background influence, focusing more on the motion changes.

Representation Learning Self-Supervised Learning

Self-supervised Temporal Discriminative Learning for Video Representation Learning

1 code implementation • 5 Aug 2020 • Jinpeng Wang, Yiqi Lin, Andy J. Ma, Pong C. Yuen

Without labelled data for network pretraining, temporal triplet is generated for each anchor video by using segment of the same or different time interval so as to enhance the capacity for temporal feature representation.

Action Recognition Representation Learning +1

Self-supervised learning using consistency regularization of spatio-temporal data augmentation for action recognition

1 code implementation • 5 Aug 2020 • Jinpeng Wang, Yiqi Lin, Andy J. Ma

Self-supervised learning has shown great potentials in improving the deep learning model in an unsupervised manner by constructing surrogate supervision signals directly from the unlabeled data.

Action Recognition Data Augmentation +1

Dynamic Label Graph Matching for Unsupervised Video Re-Identification

no code implementations • ICCV 2017 • Mang Ye, Andy J. Ma, Liang Zheng, Jiawei Li, P C Yuen

Label estimation is an important component in an unsupervised person re-identification (re-ID) system.

Graph Matching Unsupervised Person Re-Identification

Temporal Matrix Completion with Locally Linear Latent Factors for Medical Applications

no code implementations • 31 Oct 2016 • Frodo Kin Sun Chan, Andy J. Ma, Pong C. Yuen, Terry Cheuk-Fung Yip, Yee-Kit Tse, Vincent Wai-Sun Wong, Grace Lai-Hung Wong

Regular medical records are useful for medical practitioners to analyze and monitor patient health status especially for those with chronic disease, but such records are usually incomplete due to unpunctuality and absence of patients.

Imputation Matrix Completion +1

Deformable Distributed Multiple Detector Fusion for Multi-Person Tracking

no code implementations • 18 Dec 2015 • Andy J. Ma, Pong C. Yuen, Suchi Saria

For robustness to significant pose variations, deformable spatial relationship between detectors are learnt in our multi-person tracking system.

Multi-Cue Visual Tracking Using Robust Feature-Level Fusion Based on Joint Sparse Representation

no code implementations • CVPR 2014 • Xiangyuan Lan, Andy J. Ma, Pong C. Yuen

The use of multiple features for tracking has been proved as an effective approach because limitation of each feature could be compensated.

Visual Tracking