This paper proposes a generic methodology for semi-automatic generation of reliable position annotations for evaluating multi-camera people-trackers on large video datasets. Most of the annotation data is computed automatically, by estimating a consensus tracking result from multiple existing trackers and people detectors and classifying it as either reliable or not. A small subset of the data, composed of tracks with insufficient reliability is verified by a human using a simple binary decision task, a process faster than marking the correct person position. The proposed framework is generic and can handle additional trackers. We present results on a dataset of approximately 6 hours captured by 4 cameras, featuring a person in a holiday flat, performing activities such as walking, cooking, eating, cleaning, and watching TV. When aiming for a tracking accuracy of 60cm, 80% of all video frames are automatically annotated. The annotations for the remaining 20% of the frames were added after human verification of an automatically selected subset of data. This involved about 2.4 hours of manual labour. According to a subsequent comprehensive visual inspection to judge the annotation procedure, we found 99% of the automatically annotated frames to be correct. We provide guidelines on how to apply the proposed methodology to new datasets. We also provide an exploratory study for the multi-target case, applied on existing and new benchmark video sequences.

In this paper we present an efficient way to both compute and extract salient information from trace transform signatures to perform object identification tasks. We also present a feature selection analysis of the classical trace-transform functionals, which reveals that most of them retrieve redundant information causing misleading similarity measurements. In order to overcome this problem, we propose a set of functionals based on Laguerre polynomials that return orthonormal signatures between these functionals. In this way, each signature provides salient and non-correlated information that contributes to the description of an image object. The proposed functionals were tested considering a vehicle identification problem, outperforming the classical trace transform functionals in terms of computational complexity and identification rate.

This work presents a novel parallel technique to implement stack morphological filters for image processing. The method relies on applying the image bitwise decomposition to manipulate the grayscale image at a bit-plane level, while simple logical operations and Positive Boolean Functions (PBF's) are executed in parallel to derive the transformed bit-planes. The relationship between the bitwise and threshold decomposition is closely investigated and analysed, which lead us to derive an algorithm whose control flow is full binary encoded. Furthermore, the algorithm exhibits an interesting performance, which depends on the image histogram thanks to its hierarchical processing and the study of the relationship among binary decompositions.

A novel approach to implement gray-scale morphological operations is presented in this work. This new technique is based on the bitwise decomposition of the gray-scale image, yielding bitplanes disposed according to their bit of significance. It is of particular interest for implementations on focal plane processors. Our approach relies on the binary search method to obtain either the maximum or minimum on a local neighborhood by manipulating the binary levels resulting from the bitwise decomposition with simple logic functions. This contrasts significantly with the classical threshold decomposition (TD) approach, on which most of the current techniques are based on. Our method shows better efficiency than TD implementations. Further gains can be obtained because our method shows a strong dependency on the image dynamic range.

In this paper, a new method to compute the image histogram is presented, along with the image maximum and minimum values. It is intended for highly parallel architectures such as the ones found in focal plane processors (FPP). This new approach exploits this parallelism relying on the privatization technique to avoid the memory collision problem, while the bin frequency is obtained through image bitwise manipulation. Unlike traditional privatization techniques, our method exhibits a trade-off between processing time and bin size. That is, it can be adapted as a power-of-two bin size histogram and the computation time decreases exponentially as the bin size is reduced on each power of two, allowing high computational flexibility.