DocWire News Editors

DocwireNews

A group of researchers from UCLA Samueli School of Engineering and Stanford have recently created an artificial intelligence (AI) system that can identify and discover objects it encounters using visual learning methods similar to a human&rsquo;s.This AI system is currently in a &ldquo;computer vision&rdquo; class of technology, which allows computers to efficiently analyze visual images. This is a very important step leading to AI systems that are capable of self-learning, intuition, reasoning, and decision making in a more human-like manner. With current AI computer visual systems becoming more powerful and able as complexity increases, they are very limited to specific tasks. These systems have the ability to identify things that they have been trained and programmed to see but lack the flexibility to adapt.One skill that current AI visual systems lack is the ability to deduce a full image from only a partial image of an object- for example if you were to see an image of a leaf and a branch you could envision a whole tree from this. These systems also struggle to view objects in unfamiliar settings, with different aspects of the scene deceiving them.<a href="https://proddocwire.wpengine.com/docwire-pick/future-of-medicine-picks/researchers-develop-ai-with-human-like-hearing/">READ MORE:&nbsp;Researchers Develop AI With Human-Like Hearing</a>Computer engineers have been aiming to create AI systems that have these abilities we see as unique to humans. Breaking through this barrier would allow computer systems to see a partial image of a human standing behind a wall and understand from the visible head and hand where the rest of the body is.<img loading="lazy" decoding="async" class="size-full wp-image-4804680 aligncenter" src="https://mumcdnstorage.blob.core.windows.net/dwnews/2018/12/motorcycle.jpg" alt="ai" width="1024" height="768">In AI visual systems, the computer is not capable of making its own rational decisions or learn on its own. These systems are merely trained by thousands of images to pick up on certain cues that have been labelled as areas of interest. To overcome this challenge, the scientists devised a three-step system that was described in Proceedings of the National Academy of Sciences.First, this AI system breaks an image up into small chunks that the researchers dubbed &ldquo;viewlets&rdquo;. Second, the computer then learns how each of these viewlets fit together to create the whole image. Lastly, the system looks at the objects in the surrounding area and distinguishes whether or not this data is relevant to the primary object or if it is extraneous.To assist the AI in adapting more like a human, the researchers immersed the system into an internet replica of the environment we live in as humans.&ldquo;Fortunately, the internet provides two things that help a brain-inspired computer vision system learn the same way humans do,&rdquo; said Vwani Roychowdhury, UCLA engineering professor and principal investigator of the study. &ldquo;One is a wealth of images and videos that depict the same types of objects. The second is that these objects are shown from many perspectives&ndash;obscured, bird&rsquo;s eye, up-close&ndash;and they are placed in different kinds of environments.&rdquo;<a href="https://proddocwire.wpengine.com/docwire-pick/future-of-medicine-picks/idx-paving-way-ai-use-medicine/">READ MORE:&nbsp;IDx is Paving the Way for AI Use in Medicine</a>To create the structure for their system, the researchers collected insight from cognitive psychology and neuroscience.&ldquo;Starting as infants, we learn what something is because we see many examples of it, in many contexts,&rdquo; Roychowdhury said. &ldquo;That contextual learning is a key feature of our brains, and it helps us build robust models of objects that are part of an integrated worldview where everything is functionally connected.&rdquo;Testing their AI system on roughly 9,000 images that showed people and various objects, the researchers found that their platform was capable of building detailed models of the human body with no external guidance or labeling of the images. Successful results were yielded via images of various vehicles as well, with their system matching, or outperforming all traditional computer visual systems.<blockquote class="twitter-tweet" data-lang="en">
Electrical and Computer Engineering Prof. Vwani Roychowdhury and team&rsquo;s latest work in advancing <a href="https://twitter.com/hashtag/AI?src=hash&amp;ref_src=twsrc%5Etfw">#AI</a> <a href="https://twitter.com/hashtag/engineerchange?src=hash&amp;ref_src=twsrc%5Etfw">#engineerchange</a> <a href="https://twitter.com/hashtag/ucla?src=hash&amp;ref_src=twsrc%5Etfw">#ucla</a> <a href="https://twitter.com/PNASNews?ref_src=twsrc%5Etfw">@PNASNews</a> <a href="https://t.co/e0zV2uJnaF">https://t.co/e0zV2uJnaF</a>
&mdash; UCLA Samueli Engineering (@UCLAengineering) <a href="https://twitter.com/UCLAengineering/status/1075904372918214656?ref_src=twsrc%5Etfw">December 21, 2018</a>
</blockquote><script async src="https://platform.twitter.com/widgets.js" charset="utf-8"></script>Source: <a href="https://samueli.ucla.edu/new-ai-system-mimics-how-humans-visualize-and-identify-objects/">UCLA Samueli School of Engineering</a>

Jordana Jampel

Positive Preliminary Results From Phase 1 Trial of Dopaminergic Neuron Graft for Parkinson’s Disease

Dementia

Bradley Warady, MD

Joel M. Topf, MD, FACP

2025 Hume Award Winner Is Dedicated to Children’s Care

NKF Spring Clinical Meetings 2025

Distant Brain Metastases in Cutaneous Melanoma May Be Mitigated With Multiple Pathway Inhibitors

Melanoma

Latest News

Unique AI System Identifies Objects Via Similar Processes As Humans

A group of researchers from UCLA Samueli School of Engineering and Stanford have recently created an artificial intelligence (AI) system that can identify and discover objects ...

About Us

Editorial Policy

Advertising

Contact Us

eNewsletters

Contribute

Career Center

MashupMD

Blood Cancers Today

Cancer Nursing Today

GU Oncology Now

Urban Health Today

Bile Duct Cancer

Colorectal Cancer

Gastroesophageal Cancer

Gastric Cancer

GEP-NETs

GIST

Liver Cancer

Pancreatic Cancer

GI Oncology Now

Anemia

Deep Vein Thrombosis

Essential Thrombocythemia

Hemophilia

Immune Thrombocytopenic Purpura

Myelodysplastic Syndromes

Myelofibrosis

Myeloproliferative Neoplasms

Polycythemia Vera

Pulmonary Embolism

Sickle Cell Disease

Thrombocytopenia

Thrombophilia

Thrombosis

Thrombotic Thrombocytopenic Purpura

Venous Thromboembolism

von Willebrand Disease

Heme Today

Acute Kidney Injury

ADPKD

Anemia and Kidney Disease

Chronic Kidney Disease

COVID-19 and Kidney Disease

Diabetes and Hypertension

End-Stage Renal Disease

Gout

Hyperkalemia

IgAN

Kidney Transplantation

Metabolic Acidosis

Nephrology Times

HER2 Breast Cancer

HR Breast Cancer

Triple-Negative Breast Cancer

Genetics & Breast Cancer

Breast Cancers Today

Atherosclerotic Disease

Atrial Fibrillation

Heart Failure

Hypertension

Interventional Cardiology

Lipid Management

Practice Tips With Dr. Lichaa

Preventive Cardiology

Stroke

Cardiology

Breast Cancer

Gynecologic Cancer

Head and Neck Cancer

Skin Cancer

Oncology

Sleep

Urology

Diabetes

Ophthalmology

— Demodex Blepharitis

Neurology

Rheumatology

Gastroenterology

Pulmonology

Infectious Diseases

— HIV

Health and Wellness

Lupus

Future of Medicine

Orthopedics

Pharmacology

Career News

More

Conferences

CardioNerds

The Sleep Source With Dr. Morse

Rheumatology Rounds With Dr. Maheswaranathan

Partners

Expert Interviews

Subscribe