ACTNET: End-to-End Learning of Feature Activations and Multi-stream Aggregation for Effective Instance Image Retrieval

International Journal of Computer Vision - We propose a novel CNN architecture called ACTNET for robust instance image retrieval from large-scale datasets. Our key innovation is a learnable...     

Learnt inverse kinematics for animation synthesis

Existing work on animation synthesis can be roughly split into two approaches, those that combine segments of motion-capture data, and those that perform inverse kinematics. In this paper, we present a method for performing animation synthesis of an articulated object (e.g. human body and a dog) from a minimal set of body joint positions, following the approach of inverse kinematics. We tackle this problem from a learning perspective. Firstly, we address the need for knowledge on the physical constraints of the articulated body, so as to avoid the generation of a physically impossible poses. A common solution is to heuristically specify the kinematic constraints for the skeleton model. In this paper however, the physical constraints of the articulated body are represented using a hierarchical cluster model learnt from a motion capture database. Additionally, we shall show that the learnt model automatically captures the correlation between different joints through simultaneous modelling of their angles. We then show how this model can be utilised to perform inverse kinematics in a simple and efficient manner. Crucially, we describe how IK is carried out from a minimal set of end-effector positions. Following this, we show how this “learnt inverse kinematics” framework can be used to perform animation syntheses on different types of articulated structures. To this end, the results presented include the retargeting of a flat surface walking animation to various uneven terrains to demonstrate the synthesis of a full human body motion from the positions of only the hands, feet and torso. Additionally, we show how the same method can be applied to the animation synthesis of a dog using only its feet and torso positions.     

Problem solving through imitation

This paper presents an approach to problem solving through imitation. It introduces the Statistical and Temporal Percept Action Coupling (ST-PAC) System which statistically models the dependency between the perceptual state of the world and the resulting actions that this state should elicit. The ST-PAC system stores a sparse set of experiences provided by a teacher. These memories are stored to allow efficient recall and generalisation over novel systems states. Random exploration is also used as a fall-back “brute-force” mechanism should a recalled experience fail to solve a scenario. Statistical models are used to couple groups of percepts with similar actions and incremental learning used to incorporate new experiences into the system. The system is demonstrated within the problem domain of a children’s shape sorter puzzle. The ST-PAC system provides an emergent architecture where competence is implicitly encoded within the system. In order to train and evaluate such emergent architectures, the concept of the Complexity Chain is proposed. The Complexity Chain allows efficient structured learning in a similar fashion to that used in biological system and can also be used as a method for evaluating a cognitive system’s performance. Tests demonstrating the Complexity Chain in learning are shown in both simulated and live environments. Experimental results show that the proposed methods allowed for good generalisation and concept refinement from an initial set of sparse examples provided by a tutor.     

Viewpoint invariant exemplar-based 3D human tracking

This paper proposes a clustered exemplar-based model for performing viewpoint invariant tracking of the 3D motion of a human subject from a single camera. Each exemplar is associated with multiple view visual information of a person and the corresponding 3D skeletal pose. The visual information takes the form of contours obtained from different viewpoints around the subject. The inclusion of multi-view information is important for two reasons: viewpoint invariance; and generalisation to novel motions. Visual tracking of human motion is performed using a particle filter coupled to the dynamics of human movement represented by the exemplar-based model. Dynamics are modelled by clustering 3D skeletal motions with similar movement and encoding the flow both within and between clusters. Results of single view tracking demonstrate that the exemplar-based models incorporating dynamics generalise to viewpoint invariant tracking of novel movements.