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.

Logo Synthesis and Manipulation with Clustered Generative Adversarial Networks

Abstract: Designing a logo for a new brand is a lengthy and tedious back-and-forth process between a designer and a client. In this paper we explore to what extent machine learning can solve the creative task of the designer. For this, we build a dataset - LLD - of 600k+ logos crawled from the world wide web. Training Generative Adversarial Networks (GANs) for logo synthesis on such multi-modal data is not straightforward and results in mode collapse for some state-of-the-art methods. We propose the use of synthetic labels obtained through clustering to disentangle and stabilize GAN training, and validate this approach on CIFAR-10 and ImageNet-small to demonstrate its generality. We are able to generate a high diversity of plausible logos and demonstrate latent space exploration techniques to ease the logo design task in an interactive manner. GANs can cope with multi-modal data by means of synthetic labels achieved through clustering, and our results show the creative potential of such techniques for logo synthesis and manipulation. Our dataset and models are publicly available at https://data.vision.ee.ethz.ch/sagea/lld/.

Scale-invariant line descriptors for wide baseline matching

Abstract: In this paper we propose a method to add scale-invariance to line descriptors for wide baseline matching purposes. While finding point correspondences among different views is a well-studied problem, there still remain difficult cases where it performs poorly, such as textureless scenes, ambiguities and extreme transformations. For these cases using line segment correspondences is a valuable addition for finding sufficient matches. Our general method for adding scale-invariance to line segment descriptors consist of 5 basic rules. We apply these rules to enhance both the line descriptor described by Bay et al. [1] and the mean-standard deviation line descriptor (MSLD) proposed by Wang et al. [14]. Moreover, we examine the effect of the line descriptors when combined with the topological filtering method proposed by Bay et al. and the recent proposed graph matching strategy from K-VLD [6]. We validate the method using standard point correspondence benchmarks and more challenging new ones. Adding scale-invariance increases the accuracy when confronted with big scale changes and increases the number of inliers in the general case, both resulting in smaller calibration errors by means of RANSAC-like techniques and epipolar estimations.

Towards mass-produced building models

Abstract: Interest in the automatic production of 3D building models has increased over the last years. The reconstruction of buildings, particularly their facades, is a hard subproblem, given the large variety in their appearances and structures. This paper discusses building facade reconstruction algorithms that process single images and exploit expectations about facade composition. In particular, we make heavy use of the repetitions that tend to occur, e.g. in windows and balconies. But this is only an example of the kind of rules found in recent architectural shape grammars. We distinguish between cases without and with substantial perspective effects in the input image. The focus is on the latter case, where also some depth layering in the facade can be performed automatically. We give several examples of real building reconstructions.

Reconstructing 3D trajectories of independently moving objects using generic constraints

Abstract: The 3D reconstruction of scenes containing independently moving objects from uncalibrated monocular sequences still poses serious challenges. Even if the background and the moving objects are rigid, each reconstruction is only known up to a certain scale, which results in a one-parameter family of possible, relative trajectories per moving object with respect to the background. In order to determine a realistic solution from this family of possible trajectories, this paper proposes to exploit the increased linear coupling between camera and object translations that tends to appear at false scales. An independence criterion is formulated in the sense of true object and camera motions being minimally correlated. The increased coupling at false scales can also lead to the destruction of special properties such as planarity, periodicity, etc. of the true object motion. This provides us with a second, 'non-accidentalness' criterion for the selection of the correct motion among the one-parameter family.

Structured Output SVM Prediction of Apparent Age, Gender and Smile from Deep Features

Abstract: We propose structured output SVM for predicting the apparent age as well as gender and smile from a single face image represented by deep features. We pose the problem of apparent age estimation as an instance of the multi-class structured output SVM classifier followed by a softmax expected value refinement. The gender and smile predictions are treated as binary classification problems. The proposed solution first detects the face in the image and then extracts deep features from the cropped image around the detected face. We use a convolutional neural network with VGG-16 architecture [25] for learning deep features. The network is pretrained on the ImageNet [24] database and then fine-tuned on IMDB-WIKI [21] and ChaLearn 2015 LAP datasets [8]. We validate our methods on the ChaLearn 2016 LAP dataset [9]. Our structured output SVMs are trained solely on ChaLearn 2016 LAP data. We achieve excellent results for both apparent age prediction and gender and smile classification.

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.