Luis Antón-Canalís

Particle Swarms as Video Sequence Inhabitants For Object Tracking in Computer Vision

A solution based on a swarm intelligence metaphor with a prey-predator scheme is proposed for real time object tracking in video sequences, which is a basic process in multiple computer vision tasks. Swarm predator particles fly on a Boid-like fashion over image prey pixels, using combined image features to guide individual movement rules. Object tracking emerges from interaction between predator particles and their environment. The paper includes methods description and experimental evaluations on video streams that illustrate the efficiency of swarm based methods in different vision tasks

Fast and accurate hand pose detection for human-robot interaction

Enabling natural human-robot interaction using computer vision based applications requires fast and accurate hand detection. However, previous works in this field assume different constraints, like a limitation in the number of detected gestures, because hands are highly complex objects difficult to locate. This paper presents an approach which integrates temporal coherence cues and hand detection based on wrists using a cascade classifier. With this approach, we introduce three main contributions: (1) a transparent initialization mechanism without user participation for segmenting hands independently of their gesture, (2) a larger number of detected gestures as well as a faster training phase than previous cascade classifier based methods and (3) near real-time performance for hand pose detection in video streams.

AddCanny: edge detector for video processing

In this paper, we present AddCanny, an Anisotropic Diffusion and Dynamic reformulation of the Canny edge detector. The proposal provides two modifications to classical Canny detector. The first one consists of using an anisotropic diffusion filter instead of a Gaussian filter as Canny does in order to obtain better edge detection and location. The second one is the replacement of the hysteresis step by a dynamic threshold process, in order to reduce blinking effect of edges during successive frames and, therefore, generate more stable edges in sequences. Also, a new performance measurement based on the Euclidean Distance Transform to evaluate the consistency of computed edges is proposed. The paper includes experimental evaluations with different video streams that illustrate the advantages of AddCanny compared to classical Canny edge detector.

An autonomous and user-independent hand posture recognition system for vision-based interface tasks

This paper presents a system for hand posture recognition that works with colour video streams under varying light conditions for human-machine interaction in vision-based interface tasks. No initialization of the system is required and no user dependence is involved. With this aim, we first model on-line each users skin colour from the skin cue imaging of his/her face detected by means of Viola and Jones detector. Afterwards, a second order isomorphism approach performs tracking on skin colour blob based detected hand. Also, we propose this approach as a mechanism to estimate hand transition states. Finally, evidences about hand postures are recognized by shape matching, which is carried out through a holistic similarity measure focused on the Hausdorff distance. The paper includes experimental evaluations of the recognition system for 16 different hand postures in different video streams. The results show that the system can be suitable for real-time interfaces using general purpose hardware.

Analysis of Relevant Maxima in Distance Transform. An Application to Fast Coarse Image Segmentation

The Distance Transform is a powerful tool that has been used in many computer vision tasks. In this paper, the use of relevant maxima in distance transforms medial axis is proposed as a method for fast image data reduction. These disc-shaped maxima include morphological information from the object they belong to, and because maxima are located inside homogeneous regions, they also sum up chromatic information from the pixels they represent. Thus, maxima can be used instead of single pixels in algorithms which compute relations among pixels, effectively reducing computation times. As an example, a fast method for color image segmentation is proposed, which can also be used for textured zones detection. Comparisons with mean shift segmentation algorithm are shown.

Distance maps from unthresholded magnitudes

A straightforward algorithm that computes distance maps from unthresholded magnitude values is presented, suitable for still images and video sequences. While results on binary images are similar to classic Euclidean Distance Transforms, the proposed approach does not require a binarization step. Thus, no thresholds are needed and no information is lost in intermediate classification stages. Experiments include the evaluation of segmented images using the watershed algorithm and the measurement of pixel value stability in video sequences.

Natural Interaction with a Robotic Head

Social robots are receiving much interest in the robotics community. The most important goal for such robots lies in their interaction capabilities. This work describes the robotic head CASIMIRO, designed and built with the aim of achieving interactions as natural as possible. CASIMIRO is a robot face with 11 degrees of freedom. Currently, the robot has audio-visual attention (based on omnidirectional vision and sound localization abilities), face detection, head gesture recognition, owner detection, etc. The results of interviews with people that interacted with the robot support the idea that the robot has relatively natural communication abilities, although certain aspects should be further developed.