Why it matters: The neural processing involved in visually recognizing even the simplest object in a natural environment is profound---and profoundly difficult to mimic. Neuroscientists have made broad advances in understanding the visual system, but much of the inner workings of biologically-based systems remain a mystery.
Using Graphics Processing Units (GPUs) -- the same technology video game designers use to render life-like graphics – MIT and Harvard researchers are now making progress faster than ever before. “We made a powerful computing system that delivers over hundred fold speed-ups relative to conventional methods,” said Nicolas Pinto, a PhD candidate in James DiCarlo’s lab at the McGovern Institute for Brain Research at MIT. “With this extra computational power, we can discover new vision models that traditional methods miss.” Pinto co-authored the PLoS study with David Cox of the Visual Neuroscience Group at the Rowland Institute at Harvard.
How they did it: Harnessing the processing power of dozens of high-performance NVIDIA graphics cards and PlayStation 3s gaming devices, the team designed a high-throughput screening process to tease out the best parameters for visual object recognition tasks. The resulting model outperformed a crop of state-of-the-art vision systems across a range of tests -- more accurately identifying a range of objects on random natural backgrounds with variation in position, scale, and rotation. Had the team used conventional computational tools, the one-week screening phase would have taken over two years to complete.
Next steps: The researchers say that their high-throughput approach could be applied to other areas of computer vision, such as face identification, object tracking, pedestrian detection for automotive applications, and gesture and action recognition. Moreover, as scientists better understand what components make a good artificial vision system, they can use these hints to better understand the human brain as well.
Watch how the MIT/Harvard researchers are finding a better way for computers to 'see' : http://www.rowland.harvard.edu/rjf/cox/plos_video.html
Source: Pinto N, Doukhan D, DiCarlo JJ, Cox DD. A high-throughput screening approach to good forms of biologically-inspired visual representation. PLoS Computational Biology. Nov 26 2009. Read the article here: http://www.ploscompbiol.org/doi/pcbi.1000579
Funding: National Institutes of Health, McKnight Endowment for Neuroscience, Jerry and Marge Burnett, the McGovern Institute for Brain Research at MIT, and the Rowland Institute at Harvard. Hardware support provided by the NVIDIA Corporation.
Jen Hirsch | Newswise Science News
Controlling robots with brainwaves and hand gestures
20.06.2018 | Massachusetts Institute of Technology, CSAIL
Innovative autonomous system for identifying schools of fish
20.06.2018 | IMDEA Networks Institute
In a recent publication in the renowned journal Optica, scientists of Leibniz-Institute of Photonic Technology (Leibniz IPHT) in Jena showed that they can accurately control the optical properties of liquid-core fiber lasers and therefore their spectral band width by temperature and pressure tuning.
Already last year, the researchers provided experimental proof of a new dynamic of hybrid solitons– temporally and spectrally stationary light waves resulting...
Scientists from the University of Freiburg and the University of Basel identified a master regulator for bone regeneration. Prasad Shastri, Professor of...
Moving into its fourth decade, AchemAsia is setting out for new horizons: The International Expo and Innovation Forum for Sustainable Chemical Production will take place from 21-23 May 2019 in Shanghai, China. With an updated event profile, the eleventh edition focusses on topics that are especially relevant for the Chinese process industry, putting a strong emphasis on sustainability and innovation.
Founded in 1989 as a spin-off of ACHEMA to cater to the needs of China’s then developing industry, AchemAsia has since grown into a platform where the latest...
The BMBF-funded OWICELLS project was successfully completed with a final presentation at the BMW plant in Munich. The presentation demonstrated a Li-Fi communication with a mobile robot, while the robot carried out usual production processes (welding, moving and testing parts) in a 5x5m² production cell. The robust, optical wireless transmission is based on spatial diversity; in other words, data is sent and received simultaneously by several LEDs and several photodiodes. The system can transmit data at more than 100 Mbit/s and five milliseconds latency.
Modern production technologies in the automobile industry must become more flexible in order to fulfil individual customer requirements.
An international team of scientists has discovered a new way to transfer image information through multimodal fibers with almost no distortion - even if the fiber is bent. The results of the study, to which scientist from the Leibniz-Institute of Photonic Technology Jena (Leibniz IPHT) contributed, were published on 6thJune in the highly-cited journal Physical Review Letters.
Endoscopes allow doctors to see into a patient’s body like through a keyhole. Typically, the images are transmitted via a bundle of several hundreds of optical...
13.06.2018 | Event News
08.06.2018 | Event News
05.06.2018 | Event News
22.06.2018 | Materials Sciences
22.06.2018 | Earth Sciences
22.06.2018 | Life Sciences