Abstract—This paper addresses the problem of human gait classification from a robust model (in)validation perspective. The main idea is to associate to each class of gaits a nominal model, subject to bounded uncertainty and measurement noise. In this context, the problem of recognizing an activity from a sequence of frames can be formulated as the problem of determining whether this sequence could have been generated by a given (model, uncertainty, and noise) triple. By exploiting interpolation theory, this problem can be recast into a nonconvex optimization. In order to efficiently solve it, we propose two convex relaxations, one deterministic and one stochastic. As we illustrate experimentally, these relaxations achieve over 83 percent and 86 percent success rates, respectively, even in the face of noisy data. Index Terms—Gait classification, activity recognition, model (in)validation, riskadjusted (in)validation.

The human visual system processes complex biological motion stimuli with high sensitivity and selectivity. The characterization of spatio-temporal generalization in the perception of biological motion is still a largely unresolved problem. We present an experiment that inves-tigates how the visual system responds to motion stimuli that interpolate spatio-temporally between natural biological motion patterns. Inspired by analogous studies in stationary ob-ject recognition, we generated stimuli that interpolate between natural perceptual categories by morphing. Spatio-temporal morphs between natural movement patterns were obtained with a technique that allows to calculate linear combinations of spatio-temporal patterns. The weights of such linear combinations define a linear metric space over the set of gen-erated movement patterns, so that the spatio-temporal similarity of the motion patterns Preprint submitted to Elsevier Science 6 February 2002 can be quantified. In our experiments, we found smooth and continuous variation of the categorization probabilities with the weights of the prototypes in the morphs. For bipedal locomotion patterns we could accurately predict the perceived properties of the morphs by

We present a novel representation of cyclic human locomotion based on a set of spatio-temporal curves of tracked points on the surface of a person. We start by extracting a set of continuous, phase aligned spatio-temporal curves from trajectories of random points tracked over several cycles of locomotion in a monocular video sequence. We analyze a PCA representation of a set of cyclic curves, pointing out properties of the representation which can be used for spatio-temporal alignment in tracking and recognition tasks. We model the curve distribution density by a mixture of Gaussians using expectation-maximization algorithm. For recognition, we use maximum a posteriori estimate combined with linear data adaptation. We tested the algorithms on CMU MoBo database with favourable results for the recognition of people “by walking ” from monocular video sequences captured from the side view. 1.