Segmentation and recognition using structure from motion point clouds

We propose an algorithm for semantic segmentation based on 3D point clouds derived from ego-motion. We motivate five simple cues designed to model specific patterns of motion and 3D world structure that vary with object category. We introduce features that project the 3D cues back to the 2D image plane while modeling spatial layout and context. A randomized decision forest combines many such features to achieve a coherent 2D segmentation and recognize the object categories present. Our main contribution is to show how semantic segmentation is possible based solely on motion-derived 3D world structure. Our method works well on sparse, noisy point clouds, and unlike existing approaches, does not need appearance-based descriptors. Experiments were performed on a challenging new video database containing sequences filmed from a moving car in daylight and at dusk. The results confirm that indeed, accurate segmentation and recognition are possible using only motion and 3D world structure. Further, we show that the motion-derived information complements an existing state-of-the-art appearance-based method, improving both qualitative and quantitative performance. © 2008 Springer Berlin Heidelberg.

Achieving robust face recognition from video by combining a weak photometric model and a learnt generic face invariant

In spite of over two decades of intense research, illumination and pose invariance remain prohibitively challenging aspects of face recognition for most practical applications. The objective of this work is to recognize faces using video sequences both for training and recognition input, in a realistic, unconstrained setup in which lighting, pose and user motion pattern have a wide variability and face images are of low resolution. The central contribution is an illumination invariant, which we show to be suitable for recognition from video of loosely constrained head motion. In particular there are three contributions: (i) we show how a photometric model of image formation can be combined with a statistical model of generic face appearance variation to exploit the proposed invariant and generalize in the presence of extreme illumination changes; (ii) we introduce a video sequence re-illumination algorithm to achieve fine alignment of two video sequences; and (iii) we use the smoothness of geodesically local appearance manifold structure and a robust same-identity likelihood to achieve robustness to unseen head poses. We describe a fully automatic recognition system based on the proposed method and an extensive evaluation on 323 individuals and 1474 video sequences with extreme illumination, pose and head motion variation. Our system consistently achieved a nearly perfect recognition rate (over 99.7% on all four databases). © 2012 Elsevier Ltd All rights reserved.

Innovative capabilities, entrepreneurial process and the dynamic growth of university spin-outs: A process-oriented study with evidence from China

This research aims to develop a conceptual framework in order to enquire into the dynamic growth process of University Spin-outs (hereafter referred to as USOs) in China, attempting to understand the capabilities configuration that are necessary for the dynamic growth. Based on the extant literature and empirical cases, this study attempts to address the question how do USOs in China build and configure the innovative capabilities to cope with the dynamic growth. This paper aims to contribute to the existing literature by providing a theoretical discussion of the USOs' dynamic entrepreneurial process, by investigating the interconnections between innovation problem-solving and the required configuration of innovative capabilities in four growth phases. Further, it presents a particular interest on the impact to the USOs' entrepreneurial innovation process by the integrative capabilities, in terms of knowledge integration, alliance, venture finance and venture governance. To date, studies that have investigated the dynamic development process of USOs in China and have recognized the heterogeneity of USOs in terms of capabilities that are required for rapid growth still remain sparse. Addressing this research gap will be of great interest to entrepreneurs, policy makers, and venture investors. ©2009 IEEE.

Product aesthetics: representing designer intent and consumer response

This thesis reports on the development of a conceptual framework for product aesthetics. By adopting the theoretical perspective that products are a medium of communication between designers and consumers, the nature of consumer response and designer intent is explored. By integrating a range of disparate literature within a single coherent framework, the varieties of consumer response to product visual form are illustrated. To investigate the ways in which designers intend to evoke these responses, a qualitative research study was undertaken. This primarily involved interviews with industrial designers and consumer investigators. Analysis of these interviews led to the development of a conceptual framework for designer intent which both mirrors, and integrates with, that produced for consumer response. By representing processes beyond design that are influential in determining product form, a broader contextual framework is presented within which product aesthetics is situated. In concluding the thesis, applications for this framework are discussed and future research directions are proposed.