About This Talk
Welcome to the Computer Vision in Advertising and Marketing (CVAM) workshop at ICCV 2025! We’re delighted to have you join us here in Honolulu, Hawaii for this exciting exploration of cutting-edge computer vision applications in digital advertising and marketing.
This workshop brings together researchers, practitioners, and industry experts to discuss how computer vision is transforming the way brands connect with audiences through visual content. Throughout the day, we’ll explore fundamental visual understanding tasks, marketing optimization systems, brand intelligence, responsible AI practices, creative generation techniques, and emerging technologies.
Our program covers six key topic areas:
Multimodal Data Understanding: Visual content understanding, video scene analysis, multi-modal content processing, and visual similarity and retrieval.
Marketing and Optimization: Real-time bidding optimization, dynamic creative optimization (DCO), conversion rate prediction, performance forecasting, A/B testing for visuals, and resource-light video and image analysis.
Brand Intelligence: Brand safety, visual brand consistency, logo/product detection, sentiment analysis, cross-platform tracking, and brand suitability.
Responsible AI and Privacy: Privacy-preserving visual analytics, bias mitigation, transparency, ethical considerations in advertising, live-streaming CTV content and brand safety, and user privacy protection.
Creative Generation and Enhancement: Generative AI for ad creation, style transfer, automated visual optimization, and personalized content generation.
Emerging Technologies: The latest innovations shaping the future of visual advertising and marketing.
We look forward to inspiring presentations, engaging discussions, and valuable networking opportunities throughout the day. Thank you for being here, and let’s make this a memorable and productive workshop!
Presenters
Alicja Kwasniewska, PhD
AI Lead, North America, Advertisement and Marketing, Amazon AWS
Software Architect and Data Scientist with PhD in AI, focused on image, text, voice, and other data analytics. Main interests cover biomedical applications of AI, computer vision and signal processing for resource-constrained devices. Author of 30+ whitepapers, presenter during International events and conferences.
Chad Neal
North American SA Leader, Advertising and Marketing,, Amazon AWS
Technologist with over 20 years of experience in the rapid design, development, & integration of large-scale computing projects. Proven track record for initiating, building and managing mission-critical technology initiatives (web, mobile, media, big data, IoT/embedded software, predictive analytics) primarily with F100 enterprise customers.
Subarna Tripathi
AI Research Scientist, Intel Labs
As a technologist and a manager, I lead visual algorithms research at Intel Labs. We are looking into research problems such as video analysis and generation; efficient long-form structured and multimodal learning without hitting the memory / compute bottleneck.
My additional responsibilities include overseeing Intel’s global AI academic investment portfolio. We do collaborate with like-minded researchers from academia and industry!
Maciej Szankin
Principal Solutions Architect, GenAI, SiMa.ai
Maciej Szankin is a Principal Solutions Architect for Generative AI at SiMa.ai, where he leads the development of LLM and VLM-based solutions for edge applications in robotics, automotive, and drones. Since joining SiMa in January 2025, his work has focused on enabling low-latency, high-efficiency GenAI on custom hardware.
Prior to SiMa, Maciej conducted research at Intel Labs, specializing in large language models and hardware-aware AutoML. He is also a PhD candidate, working on thermal image processing and neural network architecture optimization for real-time inference on resource-constrained devices. With over 30 peer-reviewed publications, his expertise spans generative AI, multi-objective optimization, distributed ML, model compression, and neural architecture search.
Tz-Ying (Gina) Wu
AI Research Scientist, Intel Labs
I am a research scientist at Intel Labs, working on Computer Vision and Deep Learning, especially on video and multi-modal contextual learning. I received my Ph.D. from University of California San Diego (UCSD), advised by Prof. Nuno Vasconcelos in the Statistical Visual Computing Lab. My research focused on long-tail recognition with realistic and evolving models, aiming to generate reliable predictions and adapt to continually changing distributions. Before coming to UCSD, I received my B.S. and M.S. from National Tsing Hua University (NTHU), where I worked with Prof. Min Sun on multi-modal video-sensor fusion in the Vision Science Lab. Please refer to my CV for more details.
Mateusz Ruminski
Vice President of Product, PrimeAudience
Advertising technology product leader, former management consultant, driving outcomes using state-of-the-art innovations.