Lifeboat Foundation

Object detection means all the techniques and means for detecting, identifying, and classifying objects in an image. Recently, the field of artificial intelligence has seen many advances thanks to deep learning and image processing. It is now possible to recognize images or even find objects inside an image. With deep learning, object detection has become very popular with several families of models (R-CNN, YOLO, etc.). However, most of the existing methods in the literature adapt to the training database and fail to generalize when faced with images belonging to different domains.

Although most architectures are optimized for well-known benchmarks, significant results have been achieved using CNNs for tasks particular to a certain domain. However, these domain-specific solutions are often well-tuned for a specific target dataset, starting with carefully chosen architecture and training techniques. This method of training models has the drawback of unnecessarily adapting the approaches to a particular dataset. To address this issue, a research team from Intel offers a different strategy that also serves as the foundation of the Intel® Geti™ platform: a dataset-agnostic template for object detection training made up of carefully selected and pre-trained models and a reliable training pipeline for additional training.

The authors experimented with architectures in three categories: lightweight, extremely accurate, and medium, to develop a scope of the models used for the various object detection datasets regardless of complexity and object size. Pretrained weights are employed to reach model convergence quickly and begin with high accuracy. In addition, a data augmentation operation is performed to augment images with a random crop, horizontal flip, and brightness and color distortions. Multiscale training was applied for medium and accurate models to make them more robust. Additionally, to strike a balance between accuracy and complexity, the authors empirically selected particular resolutions for each model after conducting several trials. Early stopping and the adaptive ReduceOnPlateau scheduler are also used to end training if a few epochs of training do not further improve the outcome.

Government scientists have reportedly made a major breakthrough in fusion energy, producing more energy from the reaction than it…

Summary: Researchers have developed a new compound, dubbed FexD, that can prevent and reverse inflammation in mouse models of inflammatory bowel disease.

Source: Salk Institute.

A drug developed by Salk Institute researchers acts like a master reset switch in the intestines. The compound, called FexD, has previously been found to lower cholesterol, burn fat, and ward off colorectal cancer in mice.

History in the making.

The James Webb Space Telescope launched a year ago in December, kicking off a new era for astronomy. Here are the telescope’s most remarkable images from 2022.

At number 20 on IE’s 22 best innovations of 2022, we take a look back at this intimate invention.

Can you imagine kissing someone you love long distance? As sci-fi films have showcased this is one innovation that is desired by many. In May of 2022, Carnegie Mellon University’s Future Interfaces Group released a new invention that consists of a headset that can achieve just that.

Continue reading “This virtual reality headset allows you to kiss over long distances” »

Japan’s Hayabusa2 mission returned the sample to Earth in December 2020.

Japan’s asteroid mission Hayabusa2 returned a piece of the asteroid Ryugu to Earth almost two years ago now, and that sample is still revealing valuable insights into the history of the early solar system.

A study by a group of scientists from the Institut de Physique du Globe de Paris, Université Paris Cité and CNRS1 has just revealed the isotopic composition of zinc and copper of asteroid Ryugu, a press statement reveals.

Continue reading “Asteroid sample mission reveals more about Earth’s mysterious origins” »

The technology could help make EVs more affordable as well as easier to charge.

A collaboration between researchers at the Yokohoma National University in Japan and the University of New South Wales Sydney, Australia has led to the development of a new electrode material that can be used in solid-state batteries. Thus the electrode doesn’t diminish after multiple charges and discharge cycles and can help in manufacturing durable batteries for electric vehicles.

As the world is turning over to electric modes of transportation in its bid to reduce carbon emissions, the battery has become a focal area of development.

Continue reading “A new electric car battery material could dramatically boost charging times” »

Prefixes that stand for 27th and 30th powers of 10 were required due to the frequent creation of data.

At the General Conference on Weights and Measures (CGPM) in November, measurement experts and government representatives from all over the world voted to add new prefixes to the International System of Units (SI), announced the United Kingdom’s National Physical Laboratory (NPL).

So, as an extension of the metric system, the International System of Units (SI), a global decimal system of weights and measurements, was suggested by the United Kingdom’s National Physical Laboratory.

Continue reading “New prefixes for extremely huge and small numbers announced” »

“I wouldn’t want to be on the team figuring out how to build a Jupiter-sized spacecraft, but the odds aren’t zero.”

A team of scientists believes we may be able to detect alien spacecraft flying through distant solar systems using the Laser Interferometer Gravitational-Wave Observatory (LIGO) in the US.

Though the scientists from Applied Physics, a research institute in New York, concede that gravitational wave (GW) detection is “in its infancy”, they also say it is “a sophisticated science” that could be used to “detect extra-terrestrial mega-technology,” in a paper available in pre-print server arXiv.