Self-supervised Object Motion and Depth Estimation from Video

We present a self-supervised learning framework to estimate the individual object motion and monocular depth from video. We model the object motion as a 6 degree-of-freedom rigid-body transformation. The instance segmentation mask is leveraged to introduce the information of object. Compared with methods which predict pixel-wise optical flow map to model the motion, our approach significantly reduces the number of values to be estimated. Furthermore, our system eliminates the scale ambiguity of predictions, through employing the pre-computed camera ego-motion and the left-right photometric consistency. Experiments on KITTI driving dataset demonstrate our system is capable to capture the object motion without external annotation, and contribute to the depth prediction in dynamic area. Our system outperforms earlier self-supervised approaches in terms of 3D scene flow prediction, and produces comparable results on optical flow estimation.

Divide-and-Conquer Adversarial Learning for High-Resolution Image and Video Enhancement

This paper introduces a divide-and-conquer inspired adversarial learning (DACAL) approach for photo enhancement. The key idea is to decompose the photo enhancement process into hierarchically multiple sub-problems, which can be better conquered from bottom to up. On the top level, we propose a perception-based division to learn additive and multiplicative components, required to translate a low-quality image or video into its high-quality counterpart. On the intermediate level, we use a frequency-based division with generative adversarial network (GAN) to weakly supervise the photo enhancement process. On the lower level, we design a dimension-based division that enables the GAN model to better approximates the distribution distance on multiple independent one-dimensional data to train the GAN model. While considering all three hierarchies, we develop multiscale and recurrent training approaches to optimize the image and video enhancement process in a weakly-supervised manner. Both quantitative and qualitative results clearly demonstrate that the proposed DACAL achieves the state-of-the-art performance for high-resolution image and video enhancement.

AI Benchmark: All About Deep Learning on Smartphones in 2019

The performance of mobile AI accelerators has been evolving rapidly in the past two years, nearly doubling with each new generation of SoCs. The current 4th generation of mobile NPUs is already approaching the results of CUDA-compatible Nvidia graphics cards presented not long ago, which together with the increased capabilities of mobile deep learning frameworks makes it possible to run complex and deep AI models on mobile devices. In this paper, we evaluate the performance and compare the results of all chipsets from Qualcomm, HiSilicon, Samsung, MediaTek and Unisoc that are providing hardware acceleration for AI inference. We also discuss the recent changes in the Android ML pipeline and provide an overview of the deployment of deep learning models on mobile devices. All numerical results provided in this paper can be found and are regularly updated on the official project website: http://ai-benchmark.com.

Semantic Understanding of Foggy Scenes with Purely Synthetic Data

This work addresses the problem of semantic scene understanding under foggy road conditions. Although marked progress has been made in semantic scene understanding over the recent years, it is mainly concentrated on clear weather outdoor scenes. Extending semantic segmentation methods to adverse weather conditions like fog is crucially important for outdoor applications such as self-driving cars. In this paper, we propose a novel method, which uses purely synthetic data to improve the performance on unseen real-world foggy scenes captured in the streets of Zurich and its surroundings. Our results highlight the potential and power of photo-realistic synthetic images for training and especially fine-tuning deep neural nets. Our contributions are threefold, 1) we created a purely synthetic, high-quality foggy dataset of 25,000 unique outdoor scenes, that we call Foggy Synscapes and plan to release publicly 2) we show that with this data we outperform previous approaches on real-world foggy test data 3) we show that a combination of our data and previously used data can even further improve the performance on real-world foggy data.

Talk2Nav: Long-Range Vision-and-Language Navigation in Cities

Autonomous driving models often consider the goal as fixed at the start of the ride. Yet, in practice, passengers will still want to influence the route, e.g. to pick up something along the way. In order to keep such inputs intuitive, we provide automatic way finding in cities based on verbal navigational instructions and street-view images. Our first contribution is the creation of a large-scale dataset with verbal navigation instructions. To this end, we have developed an interactive visual navigation environment based on Google Street View; we further design an annotation method to highlight mined anchor landmarks and local directions between them in order to help annotators formulate typical, human references to those. The annotation task was crowdsourced on the AMT platform, to construct a new Talk2Nav dataset with 10,714 routes. Our second contribution is a new learning method. Inspired by spatial cognition research on the mental conceptualization of navigational instructions, we introduce a soft attention mechanism defined over the segmented language instructions to jointly extract two partial instructions -- one for matching the next upcoming visual landmark and the other for matching the local directions to the next landmark. On the similar lines, we also introduce memory scheme to encode the local directional transitions. Our work takes advantage of the advance in two lines of research: mental formalization of verbal navigational instructions and training neural network agents for automatic way finding. Extensive experiments show that our method significantly outperforms previous navigation methods. For demo video, dataset and code, please refer to our \href{https://www.trace.ethz.ch/publications/2019/talk2nav/index.html}{project page}.

Convex Relaxations for Consensus and Non-Minimal Problems in 3D Vision

In this paper, we formulate a generic non-minimal solver using the existing tools of Polynomials Optimization Problems (POP) from computational algebraic geometry. The proposed method exploits the well known Shor's or Lasserre's relaxations, whose theoretical aspects are also discussed. Notably, we further exploit the POP formulation of non-minimal solver also for the generic consensus maximization problems in 3D vision. Our framework is simple and straightforward to implement, which is also supported by three diverse applications in 3D vision, namely rigid body transformation estimation, Non-Rigid Structure-from-Motion (NRSfM), and camera autocalibration. In all three cases, both non-minimal and consensus maximization are tested, which are also compared against the state-of-the-art methods. Our results are competitive to the compared methods, and are also coherent with our theoretical analysis. The main contribution of this paper is the claim that a good approximate solution for many polynomial problems involved in 3D vision can be obtained using the existing theory of numerical computational algebra. This claim leads us to reason about why many relaxed methods in 3D vision behave so well? And also allows us to offer a generic relaxed solver in a rather straightforward way. We further show that the convex relaxation of these polynomials can easily be used for maximizing consensus in a deterministic manner. We support our claim using several experiments for aforementioned three diverse problems in 3D vision.

Talk2Car: Taking Control of Your Self-Driving Car

A long-term goal of artificial intelligence is to have an agent execute commands communicated through natural language. In many cases the commands are grounded in a visual environment shared by the human who gives the command and the agent. Execution of the command then requires mapping the command into the physical visual space, after which the appropriate action can be taken. In this paper we consider the former. Or more specifically, we consider the problem in an autonomous driving setting, where a passenger requests an action that can be associated with an object found in a street scene. Our work presents the Talk2Car dataset, which is the first object referral dataset that contains commands written in natural language for self-driving cars. We provide a detailed comparison with related datasets such as ReferIt, RefCOCO, RefCOCO+, RefCOCOg, Cityscape-Ref and CLEVR-Ref. Additionally, we include a performance analysis using strong state-of-the-art models. The results show that the proposed object referral task is a challenging one for which the models show promising results but still require additional research in natural language processing, computer vision and the intersection of these fields. The dataset can be found on our website: http://macchina-ai.eu/

Extremely Weak Supervised Image-to-Image Translation for Semantic Segmentation

Recent advances in generative models and adversarial training have led to a flourishing image-to-image (I2I) translation literature. The current I2I translation approaches require training images from the two domains that are either all paired (supervised) or all unpaired (unsupervised). In practice, obtaining paired training data in sufficient quantities is often very costly and cumbersome. Therefore solutions that employ unpaired data, while less accurate, are largely preferred. In this paper, we aim to bridge the gap between supervised and unsupervised I2I translation, with application to semantic image segmentation. We build upon pix2pix and CycleGAN, state-of-the-art seminal I2I translation techniques. We propose a method to select (very few) paired training samples and achieve significant improvements in both supervised and unsupervised I2I translation settings over random selection. Further, we boost the performance by incorporating both (selected) paired and unpaired samples in the training process. Our experiments show that an extremely weak supervised I2I translation solution using only one paired training sample can achieve a quantitative performance much better than the unsupervised CycleGAN model, and comparable to that of the supervised pix2pix model trained on thousands of pairs.

Texture Underfitting for Domain Adaptation

Comprehensive semantic segmentation is one of the key components for robust scene understanding and a requirement to enable autonomous driving. Driven by large scale datasets, convolutional neural networks show impressive results on this task. However, a segmentation algorithm generalizing to various scenes and conditions would require an enormously diverse dataset, making the labour intensive data acquisition and labeling process prohibitively expensive. Under the assumption of structural similarities between segmentation maps, domain adaptation promises to resolve this challenge by transferring knowledge from existing, potentially simulated datasets to new environments where no supervision exists. While the performance of this approach is contingent on the concept that neural networks learn a high level understanding of scene structure, recent work suggests that neural networks are biased towards overfitting to texture instead of learning structural and shape information. Considering the ideas underlying semantic segmentation, we employ random image stylization to augment the training dataset and propose a training procedure that facilitates texture underfitting to improve the performance of domain adaptation. In experiments with supervised as well as unsupervised methods for the task of synthetic-to-real domain adaptation, we show that our approach outperforms conventional training methods.

Learning Filter Basis for Convolutional Neural Network Compression

Convolutional neural networks (CNNs) based solutions have achieved state-of-the-art performances for many computer vision tasks, including classification and super-resolution of images. Usually the success of these methods comes with a cost of millions of parameters due to stacking deep convolutional layers. Moreover, quite a large number of filters are also used for a single convolutional layer, which exaggerates the parameter burden of current methods. Thus, in this paper, we try to reduce the number of parameters of CNNs by learning a basis of the filters in convolutional layers. For the forward pass, the learned basis is used to approximate the original filters and then used as parameters for the convolutional layers. We validate our proposed solution for multiple CNN architectures on image classification and image super-resolution benchmarks and compare favorably to the existing state-of-the-art in terms of reduction of parameters and preservation of accuracy.