Monocular SLAM

Monocular Simultaneous Localization and Mapping

Traditional feature-based SLAM systems suffer in featureless environments. Hence, methods which use unprocessed image data directly have recently become popular. In our work, we revisit the common alternated optimization of pose and depth, by introducing graduated optimization for both in a truly simultaneous energy.

Project Page | Conference Paper


DIKT: Direct Incremental Kernel–PCA Tracker

DIKT is a robust holistic visual tracking tool that functions using only the appearance of the target in the initial frame of a given video sequence. It is an adaptive algorithm and updates its stored appearance model online, utilising newly encountered appearances of the object during tracking.

Project Page | Journal Paper | Poster

Incremental SFA

Incremental Slow Feature Analysis

Slow Feature Analysis (SFA) is a subspace learning method inspired by the human visual system; however, it is seldom seen in computer vision. Motivated by its application for unsupervised activity analysis, we develop SFA’s first implementation of online temporal video segmentation to detect episodes of motion changes.

Project Page | Journal Paper