Abstract: Action recognition in still images is a challenging problem in computer vision. To facilitate comparative evaluation independently of person detection, the standard evaluation protocol for action recognition uses an oracle person detector to obtain perfect bounding box information at both training and test time. The assumption is that, in practice, a general person detector will provide candidate bounding boxes for action recognition. In this paper, we argue that this paradigm is suboptimal and that action class labels should already be considered during the detection stage. Motivated by the observation that body pose is strongly conditioned on action class, we show that: 1) the existing state-of-the-art generic person detectors are not adequate for proposing candidate bounding boxes for action classification; 2) due to limited training examples, the direct training of action-specific person detectors is also inadequate; and 3) using only a small number of labeled action examples, the transfer learning is able to adapt an existing detector to propose higher quality bounding boxes for subsequent action classification. To the best of our knowledge, we are the first to investigate transfer learning for the task of action-specific person detection in still images. We perform extensive experiments on two benchmark data sets: 1) Stanford-40 and 2) PASCAL VOC 2012. For the action detection task (i.e., both person localization and classification of the action performed), our approach outperforms methods based on general person detection by 5.7% mean average precision (MAP) on Stanford-40 and 2.1% MAP on PASCAL VOC 2012. Our approach also significantly outperforms the state of the art with a MAP of 45.4% on Stanford-40 and 31.4% on PASCAL VOC 2012. We also evaluate our action detection approach for the task of action classification (i.e., recognizing actions without localizing them). For this task, our approach, without using any ground-truth person localization at test tim- , outperforms on both data sets state-of-the-art methods, which do use person locations.

Abstract: Computer vision technologies are at the core of different advanced driver assistance systems (ADAS) and will play a key role in oncoming autonomous vehicles too. One of the main challenges for such technologies is to perceive the driving environment, i.e. to detect and track relevant driving information in a reliable manner (e.g. pedestrians in the vehicle route, free space to drive through). Nowadays it is clear that machine learning techniques are essential for developing such a visual perception for driving. In particular, the standard working pipeline consists of collecting data (i.e. on-board images), manually annotating the data (e.g. drawing bounding boxes around pedestrians), learning a discriminative data representation taking advantage of such annotations (e.g. a deformable part-based model, a deep convolutional neural network), and then assessing the reliability of such representation with the acquired data. In the last two decades most of the research efforts focused on representation learning (first, designing descriptors and learning classifiers; later doing it end-to-end). Hence, collecting data and, especially, annotating it, is essential for learning good representations. While this has been the case from the very beginning, only after the disruptive appearance of deep convolutional neural networks that it became a serious issue due to their data hungry nature. In this context, the problem is that manual data annotation is a tiresome work prone to errors. Accordingly, in the late 00’s we initiated a research line consisting of training visual models using photo-realistic computer graphics, especially focusing on assisted and autonomous driving. In this paper, we summarize such a work and show how it has become a new tendency with increasing acceptance.

Abstract: With the massive explosion of machine learning in our day-to-day life, incremental and adaptive learning has become a major topic, crucial to keep up-to-date and improve classification models and their corresponding feature extraction processes. This paper presents a categorized overview of incremental feature extraction based on linear subspace methods which aim at incorporating new information to the already acquired knowledge without accessing previous data. Specifically, this paper focuses on those linear dimensionality reduction methods with orthogonal matrix constraints based on global loss function, due to the extensive use of their batch approaches versus other linear alternatives. Thus, we cover the approaches derived from Principal Components Analysis, Linear Discriminative Analysis and Discriminative Common Vector methods. For each basic method, its incremental approaches are differentiated according to the subspace model and matrix decomposition involved in the updating process. Besides this categorization, several updating strategies are distinguished according to the amount of data used to update and to the fact of considering a static or dynamic number of classes. Moreover, the specific role of the size/dimension ratio in each method is considered. Finally, computational complexity, experimental setup and the accuracy rates according to published results are compiled and analyzed, and an empirical evaluation is done to compare the best approach of each kind.

Abstract: Recent progress of self-supervised visual representation learning has achieved remarkable success on many challenging computer vision benchmarks. However, whether these techniques can be used for domain adaptation has not been explored. In this work, we propose a generic method for self-supervised domain adaptation, using object recognition and semantic segmentation of urban scenes as use cases. Focusing on simple pretext/auxiliary tasks (e.g. image rotation prediction), we assess different learning strategies to improve domain adaptation effectiveness by self-supervision. Additionally, we propose two complementary strategies to further boost the domain adaptation accuracy on semantic segmentation within our method, consisting of prediction layer alignment and batch normalization calibration. The experimental results show adaptation levels comparable to most studied domain adaptation methods, thus, bringing self-supervision as a new alternative for reaching domain adaptation. The code is available at this link. https://github.com/Jiaolong/self-supervised-da.

Abstract: Anticipating the intentions of vulnerable road users (VRUs) such as pedestrians and cyclists is critical for performing safe and comfortable driving maneuvers. This is the case for human driving and, thus, should be taken into account by systems providing any level of driving assistance, from advanced driver assistant systems (ADAS) to fully autonomous vehicles (AVs). In this paper, we show how the latest advances on monocular vision-based human pose estimation, i.e. those relying on deep Convolutional Neural Networks (CNNs), enable to recognize the intentions of such VRUs. In the case of cyclists, we assume that they follow traffic rules to indicate future maneuvers with arm signals. In the case of pedestrians, no indications can be assumed. Instead, we hypothesize that the walking pattern of a pedestrian allows to determine if he/she has the intention of crossing the road in the path of the ego-vehicle, so that the ego-vehicle must maneuver accordingly (e.g. slowing down or stopping). In this paper, we show how the same methodology can be used for recognizing pedestrians and cyclists' intentions. For pedestrians, we perform experiments on the JAAD dataset. For cyclists, we did not found an analogous dataset, thus, we created our own one by acquiring and annotating videos which we share with the research community. Overall, the proposed pipeline provides new state-of-the-art results on the intention recognition of VRUs.