How Active Learning Improves Nighttime Pedestrian Detection

Saturday, 11th April 2020 Posted 5 years ago in AI Everything AI by Phil Alsop

Active learning makes it possible for AI to automatically choose the right training data. An ensemble of dedicated DNNs goes through a pool of image frames, flagging frames that it finds to be confusing. These frames are then labeled and added to the training dataset. This process can improve DNN perception in difficult conditions, such as nighttime pedestrian detection.

AI - is it time for a rethink?

Claire Keelan, UK MD at Onnec, is warning that the UK will fall short of its AI ambitions and data...

AI - time for a rethink?

Claire Keelan, UK MD at Onnec, is warning that the UK will fall short of its AI ambitions and data...

Observability investments increase

Joshua Clay, Senior Director Solutions Engineering at Dynatrace, discusses the findings of the...

Work AI Institute delivering real-world lessons for successful AI adoption

Work AI leader Glean has launched the Work AI Institute, a first-of-its-kind research initiative -...

Survey reveals AI readiness gap

Nicolas Leszczynski, Principal Solutions Engineer at Riverbed Technology, discusses the findings of...

AI pacesetters likely to increase their business advantage in the near future

Rob Lay, Cisco, discusses the results from the third annual Cisco AI Readiness Index. A small but...

AI - the solution to AI data centre expansion

Jackson Vo, President of Gridmatic Retail, discusses the pressures on Texas as a rapidly developing...

AI - why inferencing matters as much as LLMs

David Hurtado, Managing Director of Quetta Data Centers, discusses the impact of AI on the data...

© 2026 - Angel Business Communications Limited

Made with by Angels