“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

Pedestrian detection is a key problem in computer vision, with several applications that have the potential to positively impact quality of life. In recent years, the number of approaches to detecting pedestrians in monocular images has grown steadily. However, multiple data sets and widely varying evaluation protocols are used, making direct comparisons difficult. To address these shortcomings, we perform an extensive evaluation of the state of the art in a unified framework. We make three primary contributions: 1) We put together a large, well-annotated, and realistic monocular pedestrian detection data set and study the statistics of the size, position, and occlusion patterns of pedestrians in urban scenes, 2) we propose a refined per-frame evaluation methodology that allows us to carry out probing and informative comparisons, including measuring performance in relation to scale and occlusion, and 3) we evaluate the performance of sixteen pretrained state-of-the-art detectors across six data sets. Our study allows us to assess the state of the art and provides a framework for gauging future efforts. Our experiments show that despite significant progress, performance still has much room for improvement. In particular, detection is disappointing at low resolutions and for partially occluded pedestrians.

/pdf/pedestrian-detection-an-evaluation-of-the-state-of-the-art-3rhn1vtfew.pdf

Pedestrian Detection: An Evaluation of the State of the Art

Person re-identification is to match pedestrian images from disjoint camera views detected by pedestrian detectors. Challenges are presented in the form of complex variations of lightings, poses, viewpoints, blurring effects, image resolutions, camera settings, occlusions and background clutter across camera views. In addition, misalignment introduced by the pedestrian detector will affect most existing person re-identification methods that use manually cropped pedestrian images and assume perfect detection. In this paper, we propose a novel filter pairing neural network (FPNN) to jointly handle misalignment, photometric and geometric transforms, occlusions and background clutter. All the key components are jointly optimized to maximize the strength of each component when cooperating with others. In contrast to existing works that use handcrafted features, our method automatically learns features optimal for the re-identification task from data. The learned filter pairs encode photometric transforms. Its deep architecture makes it possible to model a mixture of complex photometric and geometric transforms. We build the largest benchmark re-id dataset with 13, 164 images of 1, 360 pedestrians. Unlike existing datasets, which only provide manually cropped pedestrian images, our dataset provides automatically detected bounding boxes for evaluation close to practical applications. Our neural network significantly outperforms state-of-the-art methods on this dataset.

/pdf/deepreid-deep-filter-pairing-neural-network-for-person-re-3s2bomwgvu.pdf

DeepReID: Deep Filter Pairing Neural Network for Person Re-identification

Simultaneous tracking of multiple persons in real-world environments is an active research field and several approaches have been proposed, based on a variety of features and algorithms. Recently, there has been a growing interest in organizing systematic evaluations to compare the various techniques. Unfortunately, the lack of common metrics for measuring the performance of multiple object trackers still makes it hard to compare their results. In this work, we introduce two intuitive and general metrics to allow for objective comparison of tracker characteristics, focusing on their precision in estimating object locations, their accuracy in recognizing object configurations and their ability to consistently label objects over time. These metrics have been extensively used in two large-scale international evaluations, the 2006 and 2007 CLEAR evaluations, to measure and compare the performance of multiple object trackers for a wide variety of tracking tasks. Selected performance results are presented and the advantages and drawbacks of the presented metrics are discussed based on the experience gained during the evaluations.

/pdf/evaluating-multiple-object-tracking-performance-the-clear-tm8230vs80.pdf

Evaluating multiple object tracking performance: the CLEAR MOT metrics

国際生活機能分類(International Classification of Functioning, Disability and Health: ICF)にみた福祉・保健・医療の専門職協働における連携に関する貢献と課題

We present a system which takes as input a video stream obtained from an airborne moving platform and produces an analysis of the behavior of the moving objects in the scene. To achieve this functionality, our system relies on two modular blocks. The first one detects and tracks moving regions in the sequence. It uses a set of features at multiple scales to stabilize the image sequence, that is, to compensate for the motion of the observer, then extracts regions with residual motion and uses an attribute graph representation to infer their trajectories. The second module takes as input these trajectories, together with user-provided information in the form of geospatial context and goal context to instantiate likely scenarios. We present details of the system, together with results on a number of real video sequences and also provide a quantitative analysis of the results.

Event detection and analysis from video streams

https://www-sop.inria.fr/members/Francois.Bremond/Postscript/hongeng_cviu03.pdf

Video-based event recognition: activity representation and probabilistic recognition methods

In many surveillance systems there is a requirement todetermine whether a given person of interest has alreadybeen observed over a network of cameras. This is the personre-identification problem. The human appearance obtainedin one camera is usually different from the ones obtained inanother camera. In order to re-identify people the humansignature should handle difference in illumination, pose andcamera parameters. We propose a new appearance modelbased on spatial covariance regions extracted from humanbody parts. The new spatial pyramid scheme is applied tocapture the correlation between human body parts in orderto obtain a discriminative human signature. The humanbody parts are automatically detected using Histograms ofOriented Gradients (HOG). The method is evaluated usingbenchmark video sequences from i-LIDS Multiple-CameraTracking Scenario data set. The re-identification performanceis presented using the cumulative matching characteristic(CMC) curve. Finally, we show that the proposedapproach outperforms state of the art methods.

/pdf/person-re-identification-using-spatial-covariance-regions-of-39rmib96w4.pdf

Person Re-identification Using Spatial Covariance Regions of Human Body Parts

In many surveillance systems there is a requirement todetermine whether a given person of interest has alreadybeen observed over a network of cameras. This paperpresents two approaches for this person re-identificationproblem. In general the human appearance obtained in onecamera is usually different from the ones obtained in anothercamera. In order to re-identify people the human signatureshould handle difference in illumination, pose andcamera parameters. Our appearance models are based onhaar-like features and dominant color descriptors. The AdaBoostscheme is applied to both descriptors to achieve themost invariant and discriminative signature. The methodsare evaluated using benchmark video sequences with differentcamera views where people are automatically detectedusing Histograms of Oriented Gradients (HOG). The reidentificationperformance is presented using the cumulativematching characteristic (CMC) curve.

/pdf/person-re-identification-using-haar-based-and-dcd-based-tg2zemlute.pdf

Person Re-identification Using Haar-based and DCD-based Signature

This paper presents a new scenario recognition algorithm for Video Interpretation. We represent a scenario model by specifying the characters involved in the scenario, the sub-scenarios composing the scenario and the constraints combining the sub-scenarios. Various types of constraints can be used including spatio-temporal and logical constraints. In this paper, we focus on the performance of the recognition algorithm. Our goal is to propose an efficient algorithm for processing temporal constraints and combining several actors defined within the scenario. By efficient we mean that the recognition process is linear in function of the number of sub-scenarios and in most of the cases in function of the number of characters. To validate this algorithm in term of correctness, robustness and processing time in function of scenario and scene properties (e.g. number of persons in the scene), we have tested the algorithm on several videos of a bank branch and of an office, in on-line and off-line mode and on simulated data. We conclude by comparing our algorithm with the state of the art and showing how the definition of scenario models can influence the results of the real-time scenario recognition.

/pdf/automatic-video-interpretation-a-novel-algorithm-for-3m2nowg33x.pdf

Francois Bremond

Papers

Event detection and analysis from video streams

Video-based event recognition: activity representation and probabilistic recognition methods

Person Re-identification Using Spatial Covariance Regions of Human Body Parts

Person Re-identification Using Haar-based and DCD-based Signature

Automatic video interpretation: a novel algorithm for temporal scenario recognition