|
|
Jiaolong Xu, Peng Wang, Heng Yang and Antonio Lopez. 2019. Training a Binary Weight Object Detector by Knowledge Transfer for Autonomous Driving. IEEE International Conference on Robotics and Automation.2379–2384.
Abstract: Autonomous driving has harsh requirements of small model size and energy efficiency, in order to enable the embedded system to achieve real-time on-board object detection. Recent deep convolutional neural network based object detectors have achieved state-of-the-art accuracy. However, such models are trained with numerous parameters and their high computational costs and large storage prohibit the deployment to memory and computation resource limited systems. Low-precision neural networks are popular techniques for reducing the computation requirements and memory footprint. Among them, binary weight neural network (BWN) is the extreme case which quantizes the float-point into just bit. BWNs are difficult to train and suffer from accuracy deprecation due to the extreme low-bit representation. To address this problem, we propose a knowledge transfer (KT) method to aid the training of BWN using a full-precision teacher network. We built DarkNet-and MobileNet-based binary weight YOLO-v2 detectors and conduct experiments on KITTI benchmark for car, pedestrian and cyclist detection. The experimental results show that the proposed method maintains high detection accuracy while reducing the model size of DarkNet-YOLO from 257 MB to 8.8 MB and MobileNet-YOLO from 193 MB to 7.9 MB.
|
|
|
|
Marçal Rusiñol, David Aldavert, Ricardo Toledo and Josep Llados. 2015. Towards Query-by-Speech Handwritten Keyword Spotting. 13th International Conference on Document Analysis and Recognition ICDAR2015.501–505.
Abstract: In this paper, we present a new querying paradigm for handwritten keyword spotting. We propose to represent handwritten word images both by visual and audio representations, enabling a query-by-speech keyword spotting system. The two representations are merged together and projected to a common sub-space in the training phase. This transform allows to, given a spoken query, retrieve word instances that were only represented by the visual modality. In addition, the same method can be used backwards at no additional cost to produce a handwritten text-tospeech system. We present our first results on this new querying mechanism using synthetic voices over the George Washington
dataset.
|
|
|
|
Cristina Cañero and 16 others. 1999. Three-dimensional reconstruction and quantification of the coronary tree using intravascular ultrasound images. Proceedings of International Conference on Computer in Cardiology (CIC´99).
Abstract: In this paper we propose a new Computer Vision technique to reconstruct the vascular wall in space using a deformable model-based technique and compounding methods, based in biplane angiography and intravascular ultrasound data jicsion. It is also proposed a generalpurpose three-dimensional guided interpolation method. The three dimensional centerline of the vessel is reconstructed from geometrically corrected biplane angiographies using automatic segmentation methods and snakes. The IVUS image planes are located in the threedimensional space and correctly oriented. A led interpolation method based in B-SurJaces and snakes isused to fill the gaps among image planes
|
|
|
|
German Ros, Laura Sellart, Joanna Materzynska, David Vazquez and Antonio Lopez. 2016. The SYNTHIA Dataset: A Large Collection of Synthetic Images for Semantic Segmentation of Urban Scenes. 29th IEEE Conference on Computer Vision and Pattern Recognition.3234–3243.
Abstract: Vision-based semantic segmentation in urban scenarios is a key functionality for autonomous driving. The irruption of deep convolutional neural networks (DCNNs) allows to foresee obtaining reliable classifiers to perform such a visual task. However, DCNNs require to learn many parameters from raw images; thus, having a sufficient amount of diversified images with this class annotations is needed. These annotations are obtained by a human cumbersome labour specially challenging for semantic segmentation, since pixel-level annotations are required. In this paper, we propose to use a virtual world for automatically generating realistic synthetic images with pixel-level annotations. Then, we address the question of how useful can be such data for the task of semantic segmentation; in particular, when using a DCNN paradigm. In order to answer this question we have generated a synthetic diversified collection of urban images, named SynthCity, with automatically generated class annotations. We use SynthCity in combination with publicly available real-world urban images with manually provided annotations. Then, we conduct experiments on a DCNN setting that show how the inclusion of SynthCity in the training stage significantly improves the performance of the semantic segmentation task
Keywords: Domain Adaptation; Autonomous Driving; Virtual Data; Semantic Segmentation
|
|
|
|
Simon Jégou, Michal Drozdzal, David Vazquez, Adriana Romero and Yoshua Bengio. 2017. The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation. IEEE Conference on Computer Vision and Pattern Recognition Workshops.
Abstract: State-of-the-art approaches for semantic image segmentation are built on Convolutional Neural Networks (CNNs). The typical segmentation architecture is composed of (a) a downsampling path responsible for extracting coarse semantic features, followed by (b) an upsampling path trained to recover the input image resolution at the output of the model and, optionally, (c) a post-processing module (e.g. Conditional Random Fields) to refine the model predictions.
Recently, a new CNN architecture, Densely Connected Convolutional Networks (DenseNets), has shown excellent results on image classification tasks. The idea of DenseNets is based on the observation that if each layer is directly connected to every other layer in a feed-forward fashion then the network will be more accurate and easier to train.
In this paper, we extend DenseNets to deal with the problem of semantic segmentation. We achieve state-of-the-art results on urban scene benchmark datasets such as CamVid and Gatech, without any further post-processing module nor pretraining. Moreover, due to smart construction of the model, our approach has much less parameters than currently published best entries for these datasets.
Keywords: Semantic Segmentation
|
|
|
|
Arnau Ramisa, David Aldavert, Shrihari Vasudevan, Ricardo Toledo and Ramon Lopez de Mantaras. 2011. The IIIA30 MObile Robot Object Recognition Datset. 11th Portuguese Robotics Open.
Abstract: Object perception is a key feature in order to make mobile robots able to perform high-level tasks. However, research aimed at addressing the constraints and limitations encountered in a mobile robotics scenario, like low image resolution, motion blur or tight computational constraints, is still very scarce. In order to facilitate future research in this direction, in this work we present an object detection and recognition dataset acquired using a mobile robotic platform. As a baseline for the dataset, we evaluated the cascade of weak classifiers object detection method from Viola and Jones.
|
|
|
|
Javad Zolfaghari Bengar and 7 others. 2019. Temporal Coherence for Active Learning in Videos. IEEE International Conference on Computer Vision Workshops.914–923.
Abstract: Autonomous driving systems require huge amounts of data to train. Manual annotation of this data is time-consuming and prohibitively expensive since it involves human resources. Therefore, active learning emerged as an alternative to ease this effort and to make data annotation more manageable. In this paper, we introduce a novel active learning approach for object detection in videos by exploiting temporal coherence. Our active learning criterion is based on the estimated number of errors in terms of false positives and false negatives. The detections obtained by the object detector are used to define the nodes of a graph and tracked forward and backward to temporally link the nodes. Minimizing an energy function defined on this graphical model provides estimates of both false positives and false negatives. Additionally, we introduce a synthetic video dataset, called SYNTHIA-AL, specially designed to evaluate active learning for video object detection in road scenes. Finally, we show that our approach outperforms active learning baselines tested on two datasets.
|
|
|
|
Joan Serrat, Ferran Diego, Felipe Lumbreras and Jose Manuel Alvarez. 2007. Synchronization of Video Sequences from Free-moving Cameras. In J. Marti et al., ed. 3rd Iberian Conference on Pattern Recognition and Image Analysis.620–627. (LNCS.)
|
|
|
|
Cesar de Souza, Adrien Gaidon, Eleonora Vig and Antonio Lopez. 2016. Sympathy for the Details: Dense Trajectories and Hybrid Classification Architectures for Action Recognition. 14th European Conference on Computer Vision.697–716. (LNCS.)
Abstract: Action recognition in videos is a challenging task due to the complexity of the spatio-temporal patterns to model and the difficulty to acquire and learn on large quantities of video data. Deep learning, although a breakthrough for image classification and showing promise for videos, has still not clearly superseded action recognition methods using hand-crafted features, even when training on massive datasets. In this paper, we introduce hybrid video classification architectures based on carefully designed unsupervised representations of hand-crafted spatio-temporal features classified by supervised deep networks. As we show in our experiments on five popular benchmarks for action recognition, our hybrid model combines the best of both worlds: it is data efficient (trained on 150 to 10000 short clips) and yet improves significantly on the state of the art, including recent deep models trained on millions of manually labelled images and videos.
|
|
|
|
David Geronimo, Frederic Lerasle and Antonio Lopez. 2012. State-driven particle filter for multi-person tracking. In J. Blanc-Talon et al., ed. 11th International Conference on Advanced Concepts for Intelligent Vision Systems. Heidelberg, Springer, 467–478.
Abstract: Multi-person tracking can be exploited in applications such as driver assistance, surveillance, multimedia and human-robot interaction. With the help of human detectors, particle filters offer a robust method able to filter noisy detections and provide temporal coherence. However, some traditional problems such as occlusions with other targets or the scene, temporal drifting or even the lost targets detection are rarely considered, making the systems performance decrease. Some authors propose to overcome these problems using heuristics not explained
and formalized in the papers, for instance by defining exceptions to the model updating depending on tracks overlapping. In this paper we propose to formalize these events by the use of a state-graph, defining the current state of the track (e.g., potential , tracked, occluded or lost) and the transitions between states in an explicit way. This approach has the advantage of linking track actions such as the online underlying models updating, which gives flexibility to the system. It provides an explicit representation to adapt the multiple parallel trackers depending on the context, i.e., each track can make use of a specific filtering strategy, dynamic model, number of particles, etc. depending on its state. We implement this technique in a single-camera multi-person tracker and test
it in public video sequences.
Keywords: human tracking
|
|
