"cv" entries

Four short links: 10 June 2015

Four short links: 10 June 2015

Product Sins, Container Satire, Dong Detection, and Evolving Code Designs

  1. The 11 Deadly Sins of Product Development (O’Reilly Radar) — they’re traps that are easy to fall into.
  2. It’s the Future — satire, but like all good satire it’s built on a rich vein of truth. Genuine guffaw funny, but Caution: Contains Rude Words.
  3. Difficulty of Dong Detection — accessible piece about how automated “inappropriate” detection remains elusive. (via Mind Hacks)
  4. Evolution of Code Design at Facebook — you may not have Facebook-scale scale problems, but if you’re having scale problems then Facebook’s evolution (not just their solutions) will interest you.
Comment
Four short links: 5 June 2015

Four short links: 5 June 2015

IoT and New Hardware Movement, OpenCV 3, FBI vs Crypto, and Transactional Datastore

  1. New Hardware and the Internet of Things (Jon Bruner) — The Internet of Things and the new hardware movement are not the same thing. The new hardware movement is driven by new tools for: Prototyping (inexpensive 3D printers, CNC machine tools, cheap and powerful microcontrollers, high-level programming languages on embedded systems); Fundraising and business development (Highway1, Lab IX); Manufacturing (PCH, Seeed); Marketing (Etsy, Quirky). The IoT is driven by: Ubiquitous connectivity; Cheap hardware (i.e., the new hardware movement); Inexpensive data processing and machine learning.
  2. OpenCV 3.0 Released — I hadn’t realised how much hardware acceleration comes out of the box with OpenCV.
  3. FBI: Companies Should Help us Prevent Encryption (WaPo) — as Mike Loukides says, we are in a Post-Modern age where we don’t trust our computers and they don’t trust us. It’s jarring to hear the organisation that (over-zealously!) investigates computer crime arguing that citizens should not be able to secure their communications. It’s like police arguing against locks.
  4. cockroacha scalable, geo-replicated, transactional datastore. The Wired piece about it drops the factoid that the creators of GIMP worked on Google’s massive BigTable-successor, Colossus. From Photoshop-alike to massive file systems. Love it.
Comment
Four short links: 20 May 2015

Four short links: 20 May 2015

Robots and Shadow Work, Time Lapse Mining, CS Papers, and Software for Reproducibility

  1. Rise of the Robots and Shadow Work (NY Times) — In “Rise of the Robots,” Ford argues that a society based on luxury consumption by a tiny elite is not economically viable. More to the point, it is not biologically viable. Humans, unlike robots, need food, health care and the sense of usefulness often supplied by jobs or other forms of work. Two thought-provoking and related books about the potential futures as a result of technology-driven change.
  2. Time Lapse Mining from Internet Photos (PDF) — First, we cluster 86 million photos into landmarks and popular viewpoints. Then, we sort the photos by date and warp each photo onto a common viewpoint. Finally, we stabilize the appearance of the sequence to compensate for lighting effects and minimize flicker. Our resulting time-lapses show diverse changes in the world’s most popular sites, like glaciers shrinking, skyscrapers being constructed, and waterfalls changing course.
  3. Git Repository of CS PapersThe intention here is to both provide myself with backups and easy access to papers, while also collecting a repository of links so that people can always find the paper they are looking for. Pull the repo and you’ll never be short of airplane/bedtime reading.
  4. Software For Reproducible ScienceThis quality is indeed central to doing science with code. What good is a data analysis pipeline if it crashes when I fiddle with the data? How can I draw conclusions from simulations if I cannot change their parameters? As soon as I need trust in code supporting a scientific finding, I find myself tinkering with its input, and often breaking it. Good scientific code is code that can be reused, that can lead to large-scale experiments validating its underlying assumptions.
Comment
Four short links: 11 May 2015

Four short links: 11 May 2015

Age of Infrastructure, Facial Expressions, Proof Assistants, and Programmer Talent

  1. Welcome to the Age of Infrastructure (Annalee Newitz) — The Internet isn’t that thing in there, inside your little glowing box. It’s in your washing machine, kitchen appliances, pet feeder, your internal organs, your car, your streets, the very walls of your house. You use your wearable to interface with the world out there.
  2. Facial Performance Sensing Head-Mounted Display (YouTube) — glorious use of an Oculus headset, to capture (for reproduction on an avatar) fine-grained facial expressions. From SIGGRAPH 2015.
  3. Mathematical Proof Assistants — human augmentation in mathematics.
  4. The Programmer Talent Myth (LWN) — Jacob Kaplan-Moss on the distribution of programmer talent and the damage that the bimodal myth causes.
Comments: 3
Four short links: 12 March 2015

Four short links: 12 March 2015

Billion Node Graphs, Asynchronous Systems, Deep Learning Hardware, and Vision Resources

  1. Mining Billion Node Graphs: Patterns and Scalable Algorithms (PDF) — slides from a CMU academic’s talk at C-BIG 2012.
  2. There Is No NowOne of the most important results in the theory of distributed systems is an impossibility result, showing one of the limits of the ability to build systems that work in a world where things can fail. This is generally referred to as the FLP result, named for its authors, Fischer, Lynch, and Paterson. Their work, which won the 2001 Dijkstra Prize for the most influential paper in distributed computing, showed conclusively that some computational problems that are achievable in a “synchronous” model in which hosts have identical or shared clocks are impossible under a weaker, asynchronous system model.
  3. Deep Learning Hardware GuideOne of the worst things you can do when building a deep learning system is to waste money on hardware that is unnecessary. Here I will guide you step by step through the hardware you will need for a cheap high performance system.
  4. Awesome Computer Vision — curated list of computer vision resources.
Comment
Four short links: 22 December 2014

Four short links: 22 December 2014

Manufacturers and Consumers, Time Management, Ethical Decisions, and Faux Faces

  1. Manufacturers and Consumers (Matt Webb) — manufacturers never spoke to consumers before. They spoke with distributors and retailers. But now products are connected to the Internet, manufacturers suddenly have a relationship with the consumer. And they literally don’t know what to do.
  2. Calendar Hacks (Etsy) — inspiration for your New Year’s resolution to waste less time.
  3. Making an Ethical Decision — there actually is an [web] app for that.
  4. Masks That Look Human to Computers — an artist creates masks that look like faces to face-recognition algorithms, but not necessarily to us. cf Deep Neural Networks are Easily Fooled.
Comment: 1
Four short links: 24 November 2014

Four short links: 24 November 2014

Magic Leap, Constant Improvement, Philanthropofallacies, and Chinese Manufacturing

  1. How Magic Leap is Secretly Creating a New Alternate Reality (Gizmodo) — amazing piece of investigative tech journalism.
  2. Better All The Time (New Yorker) — What we’re seeing is, in part, the mainstreaming of excellent habits. […] Everyone works hard. Everyone is really good.
  3. Stop Trying to Save the World (New Republic) — What I want to talk shit on is the paradigm of the Big Idea—that once we identify the correct one, we can simply unfurl it on the entire developing world like a picnic blanket. (note: some pottymouth language in this article, and some analysis I wholeheartedly agree with.)
  4. Christmas in YiwuWe travelled by container ship across the East China Sea before following the electronics supply chain around China, visiting factories, distributors, wholesalers and refineries. Fascinating! 22km of corridors in the mall that dollar store buyers visit to fill their shelves. I had never seen so many variations of the same product. Dozens of Christmas stockings bearing slightly different Santas and snowmen. Small tweaks on each theme. An in-house designer creates these designs. It feels like a brute force approach to design, creating every single possibility and then letting the market decide which it wants to buy. If none of the existing designs appeal to a buyer they can get their own designs manufactured instead. When a custom design is successful, with the customer placing a large order, it is copied by the factory and offered in their range to future buyers. The factory sales agent indicated that designs weren’t protected and could be copied freely, as long as trademarks were removed. Parallels with web design left as exercise to the reader. (via the ever-discerning Mr Webb)
Comment
Four short links: 18 November 2014

Four short links: 18 November 2014

A Worm Mind Forever LEGO Voyaging, Automatic Caption Generator, ELK Stack, and Amazonian Deployment

  1. A Worm’s Mind in a Lego Body — the c. elegans worm’s 302 neurons has been sequenced, modelled in open source code, and now hooked up to a Lego robot. It is claimed that the robot behaved in ways that are similar to observed C. elegans. Stimulation of the nose stopped forward motion. Touching the anterior and posterior touch sensors made the robot move forward and back accordingly. Stimulating the food sensor made the robot move forward. There is video.
  2. Show and Tell: A Neural Image Caption Generator — Google Research paper on generating captions like “Two pizzas sitting on top of a stove top oven” from a photo. Wow.
  3. Big Data with the ELK Stack — ElasticSearch, logstash, and Kibana. Interesting and powerful combination of tools!
  4. Apollo: Amazon’s Deployment EngineApollo will stripe the rolling update to simultaneously deploy to an equivalent number of hosts in each location. This keeps the fleet balanced and maximizes redundancy in the case of any unexpected events. When the fleet scales up to handle higher load, Apollo automatically installs the latest version of the software on the newly added hosts. Lust.
Comment
Four short links: 19 August 2014

Four short links: 19 August 2014

Adjustable Ethics, Face Projection, Mod Minecraft, and Robot Grasp

  1. Robot Cars with Adjustable Ethics Settings (Wired) — no user-servicable virtues inside. In an important sense, any injury that results from our ethics setting may be premeditated if it’s foreseen.
  2. Face-Tracking with Projection Mapping: Weird (BoingBoing) — amazing video of real-time face mapping combined with projection mapping. It is, as promised, weird.
  3. LearnToModteaches you how to code by teaching you how to mod Minecraft. It gives you two different ways to code: with blocks or with Javascript. Preorder for Oct 2014 promised release. (via Wired)
  4. Grasping with Robots: Which Object is in Reach? (Robohub) — This post is part of our ongoing efforts to make the latest papers in robotics accessible to a general audience.a new approach to build a comprehensive representation of the capabilities of a robot related to reaching and grasping. Very short, very readable, as promised.
Comment: 1
Four short links: 14 August 2014

Four short links: 14 August 2014

Ceramic 3D Printing, Robo Proofs, Microservice Fail, and Amazing Graphics Tweaks

  1. $700 Ceramic-Spitting 3D Printer (Make Magazine) — ceramic printing is super interesting, not least because it doesn’t fill the world with plastic glitchy bobbleheads.
  2. Mathematics in the Age of the Turing Machine (Arxiv) — a survey of mathematical proofs that rely on computer calculations and formal proofs. (via Victoria Stodden)
  3. Failing at Microservices — deconstructed a failed stab at microservices. Category three engineers also presented a significant problem to our implementation. In many cases, these engineers implemented services incorrectly; in one example, an engineer had literally wrapped and hosted one microservice within another because he didn’t understand how the services were supposed to communicate if they were in separate processes (or on separate machines). These engineers also had a tough time understanding how services should be tested, deployed, and monitored because they were so used to the traditional “throw the service over the fence”to an admin approach to deployment. This basically lead to huge amounts of churn and loss of productivity.
  4. Transient Attributes for High-Level Understanding and Editing of Outdoor Scenes — computer vision doing more amazing things: annotate scenes (e.g., sunsets, seasons), train, then be able to adjust images. Tweak how much sunset there is in your pic? Wow.
Comment