Detecting and reading text from natural images is a hard computer vision task that is central to a variety of emerging applications. Related problems like document character recognition have been widely studied by computer vision and machine learning researchers and are virtually solved for practical applications like reading handwritten digits. Reliably recognizing characters in more complex scenes like photographs, however, is far more difficult: the best existing methods lag well behind human performance on the same tasks. In this paper we attack the problem of recognizing digits in a real application using unsupervised feature learning methods: reading house numbers from street level photos. To this end, we introduce a new benchmark dataset for research use containing over 600,000 labeled digits cropped from Street View images. We then demonstrate the difficulty of recognizing these digits when the problem is approached with hand-designed features. Finally, we employ variants of two recently proposed unsupervised feature learning methods and find that they are convincingly superior on our benchmarks.

/pdf/reading-digits-in-natural-images-with-unsupervised-feature-3ckuq1665o.pdf

Reading Digits in Natural Images with Unsupervised Feature Learning

From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

IEEE Transactions on Pattern Analysis and Machine Intelligence

/pdf/a-break-in-the-clouds-towards-a-cloud-definition-svovsw1r5a.pdf

A Break in the Clouds: Towards a Cloud Definition

Robust detection and tracking of objects is crucial for the deployment of autonomous vehicle technology. Image based benchmark datasets have driven development in computer vision tasks such as object detection, tracking and segmentation of agents in the environment. Most autonomous vehicles, however, carry a combination of cameras and range sensors such as lidar and radar. As machine learning based methods for detection and tracking become more prevalent, there is a need to train and evaluate such methods on datasets containing range sensor data along with images. In this work we present nuTonomy scenes (nuScenes), the first dataset to carry the full autonomous vehicle sensor suite: 6 cameras, 5 radars and 1 lidar, all with full 360 degree field of view. nuScenes comprises 1000 scenes, each 20s long and fully annotated with 3D bounding boxes for 23 classes and 8 attributes. It has 7x as many annotations and 100x as many images as the pioneering KITTI dataset. We define novel 3D detection and tracking metrics. We also provide careful dataset analysis as well as baselines for lidar and image based detection and tracking. Data, development kit and more information are available online.

nuScenes: A multimodal dataset for autonomous driving

In this paper we use a “frugal innovation approach” to propose an efficient and generic solution to provide support to the deployment of IoT system in rural areas. Our proposal includes an MQTT (Message Queuing Telemetry Transport) proxy to integrate generic low-cost and low-power sensor devices in a messaging system based on LoRa (Long Range) technology. MQTT allows these data to be provided to external “data lakes” so that they can be used for tasks such as reporting, visualization, advanced analytic, and machine learning. LoRa technology provides long wireless links that can be used to connect villages and towns.

Integrating an MQTT Proxy in a LoRa-Based Messaging System for Generic Sensor Data Collection.

We discuss how demand and supply of broadband communication services can find each other in geographical areas where most telecommunication operators see too high risks and too slow return on investments. We focus on rural areas of developing regions where the general level of education is low and competence in the area of communication systems is a bottleneck. We argue that communication infrastructure is as important for society as roads, power, water and sanitation. We also argue that a knowledge society will develop faster where there are open policies and regulatory regimes in the communication sector and multi-stakeholder partnerships including what business models of traditional operators regard as externalities during a market development phase. Our approach is to target basic public sector services, such as education, health-care, local government and support to local entrepreneurs, to attract investments covering capital expenditures and to leave no stone unturned to add to the revenues required to cover operational expenditures. The capacity building of human resources required to populate all links in the value chains is problem-oriented and project driven and involves close cooperation between user communities, universities, local entrepreneurs, industrial and public partners in four basic dimensions: usage, technical solutions, entrepreneurship and impact analysis, using local tertiary level education and research institutions as a neutral platform. The purpose of the workshop is to inspire others to initiate similar stakeholder partnerships and join a knowledge society development and capacity building network.

https://mirror.unhabitat.org/downloads/docs/web4dev/Workshop12ScalableBroadband.pdf

Scalable Models for Establishment of Sustainable Broadband Services in Rural Areas of developing regions

Traditional radio planning tools present a steep learning curve. We present BotRf, a Telegram Bot that facilitates the process by guiding non-experts in assessing the feasibility of radio links. Built on open source tools, BotRf can run on any smartphone or PC running Telegram. Using it on a smartphone has the added value that the Bot can leverage the internal GPS to enter coordinates. BotRf can be used in environments with low bandwidth as the generated data traffic is quite limited. We present examples of its use in Venezuela.

Marco Zennaro

Papers

Integrating an MQTT Proxy in a LoRa-Based Messaging System for Generic Sensor Data Collection.

Scalable Models for Establishment of Sustainable Broadband Services in Rural Areas of developing regions

Radio Link Planning made easy with a Telegram Bot