Luc Van Gool

Bio: Luc Van Gool is an academic researcher from Katholieke Universiteit Leuven. The author has contributed to research in topics: Computer science & Object detection. The author has an hindex of 133, co-authored 1307 publications receiving 107743 citations. Previous affiliations of Luc Van Gool include Microsoft & ETH Zurich.

Improving Point Cloud Semantic Segmentation by Learning 3D Object Detection

Abstract: Point cloud semantic segmentation plays an essential role in autonomous driving, providing vital information about drivable surfaces and nearby objects that can aid higher level tasks such as path planning and collision avoidance. While current 3D semantic segmentation networks focus on convolutional architectures that perform great for well represented classes, they show a significant drop in performance for underrepresented classes that share similar geometric features. We propose a novel Detection Aware 3D Semantic Segmentation (DASS) framework that explicitly leverages localization features from an auxiliary 3D object detection task. By utilizing multitask training, the shared feature representation of the network is guided to be aware of per class detection features that aid tackling the differentiation of geometrically similar classes. We additionally provide a pipeline that uses DASS to generate high recall proposals for existing 2-stage detectors and demonstrate that the added supervisory signal can be used to improve 3D orientation estimation capabilities. Extensive experiments on both the SemanticKITTI and KITTI object datasets show that DASS can improve 3D semantic segmentation results of geometrically similar classes up to 37.8% IoU in image FOV while maintaining high precision bird’s-eye view (BEV) detection results.

A Stereo Vision System for Support of Planetary Surface Exploration

Abstract: In this paper a system will be presented that was developed for ESA for the support of planetary exploration. The system that is sent to the planetary surface consists of a rover and a lander. The lander contains a stereo head equipped with a pan-tilt mechanism. This vision system is used both for modeling of the terrain and for localization of the rover. Both tasks are necessary for the navigation of the rover. Due to the stress that occurs during the flight a recalibration of the stereo vision system is required once it is deployed on the planet. Due to practical limitations it is infeasible to use a known calibration pattern for this purpose and therefore a new calibration procedure had to be developed that can work on images of the planetary environment. This automatic procedure recovers the relative orientation of the cameras and the pan-and tilt-axis, besides the exterior orientation for all the images. The same images are subsequently used to recover the 3D structure of the terrain. For this purpose a dense stereo matching algorithm is used that - after rectification - computes a disparity map. Finally, all the disparity maps are merged into a single digital terrain model. In this paper a simple and elegant procedure is proposed that achieves that goal. The fact that the same images can be used for both calibration and 3D reconstruction is important since in general the communication bandwidth is very limited. In addition to the use for navigation and path planning, the 3D model of the terrain is also used for Virtual Reality simulation of the mission, in which case the model is texture mapped with the original images. The system has been implemented and the first tests on the ESA planetary terrain testbed were successful.

A Geometric Approach to Lightfield Calibration

Abstract: Lightfield rendering allows fast visualization of complex scenes by view interpolation from images of densely spaced camera viewpoints. The lightfield data structure requires calibrated viewpoints, and rendering quality can be improved substantially when local scene depth is known for each viewpoint. In this contribution we propose to combine lightfield rendering with a geometry-based structure-from-motion approach that computes camera calibration and local depth estimates. The advantage of the combined approach w.r.t. a pure geometric structure recovery is that the estimated geometry need not be globally consistent but is updated locally depending on the rendering viewpoint. We concentrate on the viewpoint calibration that is computed directly from the image data by tracking image feature points. Ground-truth experiments on real lightfield sequences confirm the quality of calibration.

Predicting Pedestrian Trajectories

Abstract: Pedestrians do not walk randomly. While they move toward their desired destination, they avoid static obstacles and other pedestrians. At the same time they try not to slow down too much as well as not to speed up excessively. Studies coming from the field of social psychology show that pedestrians exhibit common behavioral patterns. For example the distance at which one individual keeps himself from others is not uniformly random, but depends on the acquaintance level of the individuals, the culture and other factors. Our goal here is to use this knowledge to build a model that probabilistically represents the future state of a pedestrian trajectory. To this end, we focus on a stochastic motion model that caters for the possible behaviors in an entire scene in a multi-hypothesis approach, using a principled modeling of uncertainties.

Vision on Conservation: Virterf

Abstract: Although more and more computer-aided technologies are utilised or demonstrated on well-known monuments or archaeological sites, there is - at least for the professionals working in the field - still some lack of clarity about their application. The main idea of our research project is to construct a prototype of a tool that tries to bring new technologies closer to the daily practice. It incorporates recent developments in computer vision and reverse engineering, while at the same time tries to answer the practical concerns from civil servants, architects, topographers, art-historians,… who are responsible for the conservation of monuments or archaeological sites. The proposed system covers automatic correspondence analysis and point cloud manipulation to build up a textured 3D model. This model then acts as the central core of a multimedia data structure for the annotation, geometric and thematic interrogation and visualisation of the building or site being studied.

Deep Residual Learning for Image Recognition

Abstract: Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

Very Deep Convolutional Networks for Large-Scale Image Recognition

Abstract: In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Very Deep Convolutional Networks for Large-Scale Image Recognition

Abstract: In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Deep Residual Learning for Image Recognition

Abstract: Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers---8x deeper than VGG nets but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

Going deeper with convolutions

Abstract: We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.