Leandro Lorenzett Dihl

Robust Adaptive Patch-Based Object Tracking Using Weighted Vector Median Filters

This paper presents a novel patch-based approach for object tracking robust to partial and short-time total occlusions. Initially, the original template is divided into rectangular sub regions (patches), and each patch is tracked independently. The displacement of the whole template is obtained using a weighted vector median filter that combines the displacement of each patch and also a predicted displacement computed based on the previous frames. An updating scheme is also applied to cope with appearance changes of the template. Experimental results indicate that the proposed scheme is robust to partial and short-time total occlusions, presenting a good compromise between accuracy and execution time when compared to other competitive approaches.

Target Tracking Using Multiple Patches and Weighted Vector Median Filters

This paper presents an approach for object tracking based on multiple disjoint patches. Initially, the target is subdivided into a set of rectangular patches, and each patch is represented parametrically by the mean vector and covariance matrix computed from a set of feature vectors that represent each pixel of the target. Each patch is tracked independently based on the Bhattacharyya distance, and the displacement of the whole template is obtained using a Weighted Vector Median Filter (WVMF), which reduces the influence of incoherently tracked patches. To smooth the obtained trajectory and also cope with short-term total occlusions, a predicted displacement vector based on the motion of the target in the previous frames is also used, and an updating scheme is applied to deal with appearance changes of the template. Experimental results indicate that the proposed scheme is robust to partial and short-time total occlusions, presenting a good compromise between accuracy and execution time when compared to other approaches.

Human upper body identification from images

Estimating human pose in static images is challenging due to the high dimensional state space, presence of image clutter and ambiguities of image observations. In this paper we propose a method to automatically segment human subjects in images, based on dominant colors, and given the face captured by a face detector. The posture is estimated using a 2D model combined with anthropometric data. Experimental results showed that the proposed technique performs well in non trivial images.

Feature-Based Face Tracking for Videoconferencing Applications

This paper proposes a new approach for face tracking based on the individual tracking of KLT features. The face is initially detected using a face detection scheme, and KLT features are distributed along the face. Each feature is tracked individually, and the displacement of the center of the face is obtained using a Weighted Vector Median Filter (WVMF) of the individual displacements. The scale change is then computed based on the position of each feature w.r.t. the center of the face. The experimental results indicate that the proposed approach is fast and robust in the presence of partial occlusions.

Procedural floor plan generation from building sketches

We present a method for automated reconstruction of building interiors from hand-drawn building sketches. Image processing is used to extract the building’s outline and openings. Then, a procedural generation algorithm creates a floor plan according to user requisites. The proposed method handles a wide variety of input image styles and building shapes, including non-convex polygons. Possible applications include architectural tools and digital content generation.

Making Them Alive

This paper presents a model to reconstruct 3D virtual humans based on a single and spontaneous image. The main goal is to use computer vision and pattern recognition techniques to build coherent virtual humans according to an input picture. To achieve this goal we provide a semi-automatic process that includes 3D posture detection, segmentation of human body parts, and silhouette processing. Such information is used to generate a 3D virtual human, which can be further animated. The approach proposed in this paper aims to speed up the creation of 3D articulated characters, providing avatars based on pictures. Experimental results indicate that our approach is a good option for generating virtual humans from images based on a few mouse clicks.

Using group behaviors to detect Hofstede cultural dimensions

This paper presents a methodology to characterize information about groups of people with the main goal of detecting cultural aspects. Based on tracked pedestrians, groups are detected and characterized. Group information is then used to find out Cultural aspects in videos, based on the Hofstede cultural dimensions theory. The presented work was tested in videos of pedestrian groups recorded in different countries and results seem promising in order to identify cultural aspects in the filmed sequences.

Detecting Crowd Features in Video Sequences

We propose a new methodology to detect social aspects of crowds in video sequences based on pedestrian features, which are obtained through image processing/computer vision techniques. The main idea is to apply and extend the concepts of Fundamental Diagram (FD) with more features, such as grouping and collectivity. Using crowd features we identify the crowd type and the main characteristics. In addition, we also investigated two further results: the visual assessment of people in real video sequences in order to detect crowd characteristics, and the usage of our method to detect similarity of crowds in videos.

Recovering 3D human pose based on biomechanical constraints, postures comfort and image shading

Abstract This paper presents a new model to identify 3D human poses in pictures, given a single input image. The proposed approach is based on a well known model found in the literature, including improvements in terms of biomechanical restrictions aiming to reduce the number of 3D possible postures that correctly represent the pose in the 2D image. Since the generated set of poses can have more than one possible posture, we propose a ranking system in order to suggest the best generated postures according to a “comfort” criterion and shading characteristics in the image as well. The comfort criterion adopts assumptions in terms of pose equilibrium, while the shading criterion eliminates the ambiguities of postures taken into account the image illumination. We must emphasize that the removal of ambiguous 3D poses related to a single image is the main focus of this work. The achieved results were analyzed w.r.t. visual inspection of users as well as a state of the art technique and indicate that our model contributed in some way to the solution of that challenge problem.

Using Big Five Personality Model to Detect Cultural Aspects in Crowds

The use of information technology in the study of human behavior is a subject of great scientific interest. Cultural and personality aspects are factors that influence how people interact with one another in a crowd. This paper presents a methodology to detect cultural characteristics of crowds in video sequences. Based on filmed sequences, pedestrians are detected, tracked and characterized. Such information is then used to find out cultural differences in those videos, based on the Big-five personality model. Regarding cultural differences of each country, results indicate that this model generates coherent information when compared to data provided in literature.