Researchers at the SketchX, University of Surrey have recently developed a meta learning-based model that allows users to retrieve images of specific items simply by sketching them on a tablet, smartphone, or on other smart devices. This framework was outlined in a paper set to be presented at the European Conference on Computer Vision (ECCV), one of the top three flagship computer vision conferences along with CVPR and ICCV.
Researchers at the SketchX, University of Surrey have recently developed a meta learning-based model that allows users to retrieve images of specific items simply by sketching them on a tablet, smartphone, or on other smart devices. This framework was outlined in a paper set to be presented at the European Conference on Computer Vision (ECCV), one of the top three flagship computer vision conferences along with CVPR and ICCV.
“This is the latest along the line of work on ‘fine-grained image retrieval,’ a problem that my research lab (SketchX, which I direct and founded back in 2012) pioneered back in 2015, with a paper published in CVPR 2015 titled ‘Sketch Me That Shoe,’” Yi-Zhe Song, one of the researchers who carried out the study, told TechXplore. “The idea behind our paper is that it is often hard or impossible to conduct image retrieval at a fine-grained level, (e.g., finding a particular type of shoe at Christmas, but not any shoe).”
In the past, some researchers tried to devise models that can retrieve images based on text or voice descriptions. Text might be easier for users to produce, yet it was found only to work at a coarse level. In other words, it can become ambiguous and ineffective when trying to describe details.
Posted in neuroscience
Summary: A new study using 3D-MINFLUX technology allowed researchers to image the active zone of synapses with the precision of a few nanometers.
Source: University of Gottingen.
Scientists at the Institute for Auditory Neuroscience, UMG, the Max Planck Institute for Multidisciplinary Sciences and the Abberior Instruments GmbH have applied high-resolution 3D-MINFLUX technology for precise 3D representation of the molecular organization in the active zone of rod photoreceptor cells.
Vulnerability researchers have found security issues in a GPS tracker that is advertised as being present in about 1.5 million vehicles in 169 countries.
A total of six vulnerabilities affect the MiCODUS MV720 device, which is present in vehicles used by several Fortune 50 firms, governments in Europe, states in the U.S., a military agency in South America, and a nuclear plant operator.
The risks stemming from the findings are significant and impact both privacy and security. A hacker compromising an MV720 device could use it for tracking or even immobilizing the vehicle carrying it, or to collect information about the routes, and manipulate data.
Penn State agricultural engineers have developed, for the first time, a prototype “end-effector” capable of deftly removing unwanted apples from trees—the first step toward robotic, green-fruit thinning.
The development is important, according to Long He, assistant professor of agricultural and biological engineering, because manual thinning is a labor-intensive task, and the shrinking labor force in apple production makes manual thinning economically infeasible. His research group in the College of Agricultural Sciences conducted a new study that led to the end-effector.
The apple crop is a high-value agricultural commodity in the U.S., with an annual total production of nearly 10 billion pounds and valued at nearly $3 billion, according to He, who is a leader in agricultural robotics research, previously developing automated components for mushroom picking and apple tree pruning. Green-fruit thinning—the process of discarding excess fruitlets in early summer, mainly to increase the remaining fruit size and quality—is one of the most important aspects of apple production.
An extraordinary image of the centre of the spiral galaxy NGC 628 taken by the James Webb Space Telescope (JWST) has drawn comparisons to a Doctor Who vortex, but it could reveal important clues about how dust behaves in galaxies.
The image is a composite of three sets of data at different wavelengths taken by JWST’s mid-infrared instrument team. Gabriel Brammer at the University of Copenhagen in Denmark, who isn’t affiliated with the team, downloaded the data and translated each of the infrared wavelengths to red, green and blue before combining them to produce one image.
NGC 628 has been imaged with visible light by other telescopes, including Hubble, and looks similar to our own Milky Way if viewed from above the galactic plane. But JWST’s ability to observe infrared light in high resolution reveals a hidden structure. “If our eyes could see in these mid-infrared wavelengths, the night sky would look a lot more like this picture, which I think would be spectacular, maybe a little terrifying,” says Brammer.
Scientists report that damage to the James Webb Space Telescope (JWST) caused by a micrometeoroid impact in late May 2022 may be more severe than previously believed.
A group of scientists evaluated the performance of the space telescope throughout its commissioning phase in a new report released last week following Webb’s amazing first photographs.
VIDEO: If the history of Earth was on a timeline from Los Angeles to New York, the American Revolution would have been fought 8 inches ago.
Researchers at Rice University have shown how they can hack the brains of fruit flies to make them remote controlled. The flies performed a specific action within a second of a command being sent to certain neurons in their brain.
The team started by genetically engineering the flies so that they expressed a certain heat-sensitive ion channel in some of their neurons. When this channel sensed heat, it would activate the neuron – in this case, that neuron caused the fly to spread its wings, which is a gesture they often use during mating.
The heat trigger came in the form of iron oxide nanoparticles injected into the insects’ brains. When a magnetic field is switched on nearby, those particles heat up, causing the neurons to fire and the fly to adopt the spread-wing pose.
Now, researchers at the Karlsruhe Institute of Technology (KIT) have developed a prototype for fully scalable all–perovskite tandem solar modules. They were able to scale up individual perovskite cells with a power conversion efficiency of up to 23.5% at an aperture area of 0.1 square centimeters to all-perovskite tandem solar modules with an efficiency of up to 19.1% with an aperture area of 12.25 square centimeters.
The aperture area is the usable part of the surface that is not covered by electrodes, frames, or fasteners. At approximately five percent, the loss of efficiency when upscaling is relatively low. “This is the first report of an all-perovskite tandem solar module worldwide,” says Dr. Bahram Abdollahi Nejand, lead author of the publication and team leader for all-perovskite tandem solar modules.
The KIT researchers claimed this remarkable result is the first of its kind reported worldwide. To obtain this, the team increased the efficiency by optimizing the light path and reducing reflections in the solar cell architecture. They implemented an efficient layout for tandem solar modules using high-throughput laser scribing that enables the production of functional tandem solar mini-modules with two-terminal interconnected cell strips. Lastly, they used coating processes (blade coating and vacuum deposition) that are already established in industrial practice.
