Science comes from intuition, and is rationalized by reasons.

Unsupervised Visual Instance Search

In this research subject, we explore instance level feature representation based on unsupervised deep learning approach. The instance level representation leads to a universal representation for various of image-related search tasks. Project shown in slides Source code on Github

Vector Database

Nearest neighbor search and k-nearest neighbor graph construction are two fundamental issues that arise from many disciplines such as multimedia information retrieval, data-mining, and machine learning. They become more and more imminent given the big data emerge in various fields in recent years. We have designed very efficient indexing structure that perform well on both the CPU and the GPU. Project shown in slides Source code on Github

Anomaly Detection and Trend Prediction on IT Operation Data

Status prediction and anomaly detection are two fundamental tasks in automatic IT systems monitoring. In our design, the variational auto-encoder (VaE) and long short-term memory (LSTM) are joined together. Project shown in slides

© 2014 - 2024 by Computing Multimedia Lab. All rights are reserved. Last updated on Aug-30-2024.