Abstract: Ground-truth data is essential for the objective evaluation of object detection methods in computer vision. Many works claim their method is robust but they support it with experiments which are not quantitatively assessed with regard some ground-truth. This is one of the main obstacles to properly evaluate and compare such methods. One of the main reasons is that creating an extensive and representative ground-truth is very time consuming, specially in the case of video sequences, where thousands of frames have to be labelled. Could such a ground-truth be generated, at least in part, automatically? Though it may seem a contradictory question, we show that this is possible for the case of video sequences recorded from a moving camera. The key idea is transferring existing frame segmentations from a reference sequence into another video sequence recorded at a different time on the same track, possibly under a different ambient lighting. We have carried out experiments on several video sequence pairs and quantitatively assessed the precision of the transformed ground-truth, which prove that our approach is not only feasible but also quite accurate.

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: Depth can complement RGB with useful cues about object volumes and scene layout. However, RGB-D image datasets are still too small for directly training deep convolutional neural networks (CNNs), in contrast to the massive monomodal RGB datasets. Previous works in RGB-D recognition typically combine two separate networks for RGB and depth data, pretrained with a large RGB dataset and then fine tuned to the respective target RGB and depth datasets. These approaches have several limitations: 1) only use low-level filters learned from RGB data, thus not being able to exploit properly depth-specific patterns, and 2) RGB and depth features are only combined at high-levels but rarely at lower-levels. In this paper, we propose a framework that leverages both knowledge acquired from large RGB datasets together with depth-specific cues learned from the limited depth data, obtaining more effective multi-source and multi-modal representations. We propose a multi-modal combination method that selects discriminative combinations of layers from the different source models and target modalities, capturing both high-level properties of the task and intrinsic low-level properties of both modalities.

Abstract: Deep networks trained on demonstrations of human driving have learned to follow roads and avoid obstacles. However, driving policies trained via imitation learning cannot be controlled at test time. A vehicle trained end-to-end to imitate an expert cannot be guided to take a specific turn at an upcoming intersection. This limits the utility of such systems. We propose to condition imitation learning on high-level command input. At test time, the learned driving policy functions as a chauffeur that handles sensorimotor coordination but continues to respond to navigational commands. We evaluate different architectures for conditional imitation learning in vision-based driving. We conduct experiments in realistic three-dimensional simulations of urban driving and on a 1/5 scale robotic truck that is trained to drive in a residential area. Both systems drive based on visual input yet remain responsive to high-level navigational commands. The supplementary video can be viewed at this https URL

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.

Abstract: A long-standing view is that lightness illusions are by-products of strategies employed by the visual system to stabilize its perceptual representation of surface reflectance against changes in illumination. Computationally, one such strategy is to infer reflectance from the retinal image, and to base the lightness percept on this inference. CNNs trained to infer reflectance from images have proven successful at solving this problem under limited conditions. To evaluate whether these CNNs provide suitable starting points for computational models of human lightness perception, we tested a state-of-the-art CNN on several lightness illusions, and compared its behaviour to prior measurements of human performance. We trained a CNN (Yu & Smith, 2019) to infer reflectance from luminance images. The network had a 30-layer hourglass architecture with skip connections. We trained the network via supervised learning on 100K images, rendered in Blender, each showing randomly placed geometric objects (surfaces, cubes, tori, etc.), with random Lambertian reflectance patterns (solid, Voronoi, or low-pass noise), under randomized point+ambient lighting. The renderer also provided the ground-truth reflectance images required for training. After training, we applied the network to several visual illusions. These included the argyle, Koffka-Adelson, snake, White’s, checkerboard assimilation, and simultaneous contrast illusions, along with their controls where appropriate. The CNN correctly predicted larger illusions in the argyle, Koffka-Adelson, and snake images than in their controls. It also correctly predicted an assimilation effect in White's illusion. It did not, however, account for the checkerboard assimilation or simultaneous contrast effects. These results are consistent with the view that at least some lightness phenomena are by-products of a rational approach to inferring stable representations of physical properties from intrinsically ambiguous retinal images. Furthermore, they suggest that CNN models may be a promising starting point for new models of human lightness perception.