"computer vision" entries

Four short links: 27 November 2015

Android Insecurity, Clear Photos, Speech to Emotion, and Microexpressions from Video

  1. 87% of Android Devices Insecure — researchers find they’re vulnerable to malicious apps because manufacturers have not provided regular security updates. (via Bruce Schneier)
  2. A Computational Approach for Obstruction-Free Photography (Google Research) — take multiple photos from different angles through occlusions like a window with raindrops or reflections, and their software will assemble an unoccluded image. (via Greg Linden)
  3. Algorithms for Affective SensingResults show that the system achieves a six-emotion decision-level correct classification rate of 80% for an acted dataset with clean speech. This PhD thesis is research into algorithm for determining emotion from speech samples, which does so more accurately than humans in a controlled test. (via New Scientist)
  4. Software Learns to Recognise Microexpressions (MIT Technology Review) — Li and co’s machine matched human ability to spot and recognize microexpressions and significantly outperformed humans at the recognition task alone.
Four short links: 20 November 2015

Four short links: 20 November 2015

Table Mining, Visual Microphones, Platformed Government, and NP-Hard Video Games

  1. DeepDive — Stanford project to create structured data (SQL tables) from unstructured information (text documents) and integrate such data with an existing structured database. DeepDive is used to extract sophisticated relationships between entities and make inferences about facts involving those entities. Code is open source (Apache v2 license). (via Infoworld)
  2. Visual Microphone (MIT) — turn everyday objects — a glass of water, a potted plant, a box of tissues, or a bag of chips — into visual microphones using high-speed photography to detect the small vibrations caused by sound. (via Infoworld)
  3. 10 Rules for Distributed/Networked/Platformed Government (Richard Pope) — Be as vigilant against creating concentrations of power as you are in creating efficiency or bad user experiences. (via Paul Downey)
  4. Classic Nintendo Games are (Computationally) HardWe prove NP-hardness results for five of Nintendo’s largest video game franchises: Mario, Donkey Kong, Legend of Zelda, Metroid, and Pokemon.
Four short links: 16 November 2015

Four short links: 16 November 2015

Hospital Hacking, Security Data Science, Javascript Face-Substitution, and Multi-Agent Systems Textbook

  1. Hospital Hacking (Bloomberg) — interesting for both lax regulation (“The FDA seems to literally be waiting for someone to be killed before they can say, ‘OK, yeah, this is something we need to worry about,’ ” Rios says.) and the extent of the problem (Last fall, analysts with TrapX Security, a firm based in San Mateo, Calif., began installing software in more than 60 hospitals to trace medical device hacks. […] After six months, TrapX concluded that all of the hospitals contained medical devices that had been infected by malware.). It may take a Vice President’s defibrillator being hacked for things to change. Or would anybody notice?
  2. Cybersecurity and Data Science — pointers to papers in different aspects of using machine learning and statistics to identify misuse and anomalies.
  3. Real-time Face Substitution in Javascript — this is awesome. Moore’s Law is amazing.
  4. Multi-Agent Systems — undergraduate textbook covering distributed systems, game theory, auctions, and more. Electronic version as well as printed book.
Four short links: 22 October 2015

Four short links: 22 October 2015

Predicting activity, systems replacement fail, Khan React style, and an interoperability system for the Web

  1. Predicting Daily Activities from Egocentric Images Using Deep LearningOur technique achieves an overall accuracy of 83.07% in predicting a person’s activity [from images taken by a camera worn all day by a person] across the 19 activity classes.
  2. Trying to Replace Multiple Systems with One Can Lead to None (IEEE) — check out that final graph, it’s a doozy. It’s a graph of x against time, from various “this project is great, it will replace x systems with 1″ claims about a single project. Software projects should come with giant warning labels: “most fail, you are about to set your money on fire. Are you sure? [Y/N/Abort/Restart]”
  3. Khan React Style Guide — in case you’re dipping your toes into the cool kids’ pool.
  4. ballistaAn interoperability system for the modern Web. Like intents.
Four short links: 20 October 2015

Four short links: 20 October 2015

HyperCam, half-arsed software development, perceptions of productivity, John McCarthy's conditional expressions

  1. HyperCam (PDF) — paper from Ubicomp 2015 on a low-cost implementation of a multispectral camera and a software approach that automatically analyzes the scene and provides a user with an optimal set of images that try to capture the salient information of the scene. Can see ripeness of fruit, and veins in hands.
  2. Manifesto for Half-Arsed Software DevelopmentResponding to change over following a plan … provided a detailed plan is in place to respond to the change, and it is followed precisely.
  3. Software Developers’ Perceptions of ProductivityIn both studies, we found that developers perceive their days as productive when they complete many or big tasks without significant interruptions or context switches. Yet, the observational data we collected shows our participants performed significant task and activity switching while still feeling productive. (via Never Work in Theory)
  4. The Language of ChoiceIn the ’50s John McCarthy invented conditional expressions. Utility computing, AI, Lisp, and now what I know as C’s ?: syntax. His legend lives on.
Comments: 2
Four short links: 1 September 2015

Four short links: 1 September 2015

People Detection, Ratings Patterns, Inspection Bias, and Cloud Filesystem

  1. End-to-End People Detection in Crowded Scenes — research paper and code. When parsing the title, bind “end-to-end” to “scenes” not “people”.
  2. Statistical Patterns in Movie Ratings (PLOSone) — We find that the distribution of votes presents scale-free behavior over several orders of magnitude, with an exponent very close to 3/2, with exponential cutoff. It is remarkable that this pattern emerges independently of movie attributes such as average rating, age and genre, with the exception of a few genres and of high-budget films.
  3. The Inspection Bias is EverywhereIn 1991, Scott Feld presented the “friendship paradox”: the observation that most people have fewer friends than their friends have. He studied real-life friends, but the same effect appears in online networks: if you choose a random Facebook user, and then choose one of their friends at random, the chance is about 80% that the friend has more friends. The friendship paradox is a form of the inspection paradox. When you choose a random user, every user is equally likely. But when you choose one of their friends, you are more likely to choose someone with a lot of friends. Specifically, someone with x friends is overrepresented by a factor of x.
  4. s3qla file system that stores all its data online using storage services like Google Storage, Amazon S3, or OpenStack. S3QL effectively provides a hard disk of dynamic, infinite capacity that can be accessed from any computer with internet access running Linux, FreeBSD or OS-X. (GPLv3)
Four short links: 13 August 2015

Four short links: 13 August 2015

Learning Style, Artisinal Cash, Docs at Scale, and Homophily Research

  1. Elements of Style: Learning Perceptual Shape Style Similarity — code and data for research that helps perceive stylistic similarity between objects that transcends structure and function. For example, we can see a common style such as “Danish modern” in both a table and chair, though they have different structures. Until now, machines have found it difficult to do the same. (That quote cribbed from the phys.org writeup) Our new AI overlords may be cruel and heartless, but they’ll be able to tell Danish Modern from Shaker.
  2. The Advent of Artisinal Cash (NY Times) — details the rise of local physical currency around the world. Nonetheless, the use of traditional paper money is clearly on the wane. Perhaps these smaller, more attractive artisanal paper notes are merely last bursts of glory before it disappears entirely. Though as Mr. Deller, the artist behind the latest Brixton pound, said, “As long as there are drug deals and criminality, there’ll be a need for cash.”
  3. Documentation at Scale1. Acknowledge that brute force doesn’t work; 2. Make documentation a first class citizen; 3. Make documentation executable; 4. Track the intent.
  4. Exposure to Ideologically Diverse Information on Facebook (Facebook Research) — Friends shared substantially less cross-cutting news from sources aligned with an opposing ideology. People encountered roughly 15% less cross-cutting content in news feeds due to algorithmic ranking and clicked through to 70% less of this cross-cutting content. Within the domain of political news encountered in social media, selective exposure appears to drive attention.

We make the software, you make the robots

An interview with Andreas Mueller, on scikit-learn and usable machine learning software.

Get notified when our free report “Evaluating Machine Learning Models: A beginner’s guide to key concepts and pitfalls,” by Alice Zheng, is available for download.


Superpixels example from Andreas Mueller’s thesis paper (PDF), used with permission.

A few weeks ago, I had the pleasure of sitting down (virtually, over Skype) with Andreas Mueller, core developer and maintainer of the popular scikit-learn machine learning library. We had previously bonded over our shared goals of making useful machine learning software, so I jumped at the chance to interview him.

Mueller wears many hats at work. He is one of the key maintainers of the popular Python machine learning library scikit-learn. Holding a doctorate in computer vision from the University of Bonn in Germany, he currently works on open science at New York University’s Center for Data Science. He speaks at conferences around the world and has a fanbase of 5,000+ followers on Twitter and about as many reputation points on Stack Overflow. In other words, this man has got mad street cred. He started out doing pure math in academia, and has now achieved software developer cult idol status. Read more…

Four short links: 31 July 2015

Four short links: 31 July 2015

Robot Swarms, Google Datacenters, VR Ecosystem, and DeepDream Visualised

  1. Buzz: An Extensible Programming Language for Self-Organizing Heterogeneous Robot Swarms (arXiv) — Swarm-based primitives allow for the dynamic management of robot teams, and for sharing information globally across the swarm. Self-organization stems from the completely decentralized mechanisms upon which the Buzz run-time platform is based. The language can be extended to add new primitives (thus supporting heterogeneous robot swarms), and its run-time platform is designed to be laid on top of other frameworks, such as Robot Operating System.
  2. Jupiter Rising: A Decade of Clos Topologies and Centralized Control in Google’s Datacenter Network (PDF) — Our datacenter networks run at dozens of sites across the planet, scaling in capacity by 100x over 10 years to more than 1Pbps of bisection bandwidth. Wow, their Wi-Fi must be AMAZING!
  3. Nokia’s VR Ambitions Could Restore Its Tech Lustre (Bloomberg) — the VR ecosystem map is super-interesting.
  4. Visualising GoogleNet Classes — fascinating to see squirrel monkeys and basset hounds emerge from nothing. It’s so tempting to say, “this is what the machine sees in its mind when it thinks of basset hounds,” even though Boring Brain says, “that’s bollocks and you know it!”
Four short links: 27 July 2015

Four short links: 27 July 2015

Google’s Borg, Georgia v. Malamud, SLAM-aware system, and SmartGPA

  1. Large-scale Cluster Management at Google with BorgGoogle’s Borg system is a cluster manager that runs hundreds of thousands of jobs, from many thousands of different applications, across a number of clusters, each with up to tens of thousands of machines. […] We present a summary of the Borg system architecture and features, important design decisions, a quantitative analysis of some of its policy decisions, and a qualitative examination of lessons learned from a decade of operational experience with it.
  2. Georgia Sues Carl Malamud (TechDirt) — for copyright infringement… for publishing an official annotated copy of the state's laws. […] the state points directly to the annotated version as the official laws of the state.
  3. Monocular SLAM Supported Object Recognition (PDF) — a monocular SLAM-aware object recognition system that is able to achieve considerably stronger recognition performance, as compared to classical object recognition systems that function on a frame-by-frame basis. (via Improving Object Recognition for Robots)
  4. SmartGPA: How Smartphones Can Assess and Predict Academic Performance of College Students (PDF) — We show that there are a number of important behavioral factors automatically inferred from smartphones that significantly correlate with term and cumulative GPA, including time series analysis of activity, conversational interaction, mobility, class attendance, studying, and partying.
Comment: 1