Schedule

All times are listed in three time zones for convenience: Seattle (PDT), Central European Summer Time (CEST), and Beijing (CST).

The program for the Vision for All Seasons workshop brings together leading researchers and practitioners working on the challenges of adverse weather and lighting conditions. Presentations throughout the day focus on how fog, rain, snow, ice, low light, nighttime, glare, and shadows degrade the performance of vision systems used in autonomous driving, surveillance, and other outdoor applications. The invited talks highlight both fundamental research directions and practical deployment considerations, bridging the gap between laboratory benchmarks and the messy realities of operating in uncontrolled environments. Attendees are encouraged to engage with speakers and each other during the live sessions, fostering discussion that extends beyond the scheduled talks.

A central theme running through the workshop program is the recognition that current computer vision algorithms, while performing impressively on favorably benchmarked datasets, undergo severe degradation when exposed to real-world adverse conditions. The invited speakers address this gap from complementary angles, including sensor design, image formation physics, and learning-based approaches that aim to make perception systems more robust. The program emphasizes that progress in this direction requires not only better algorithms but also a deeper understanding of how weather and lighting physically affect the signals captured by cameras and other sensors. This holistic perspective is essential for building systems that can operate smoothly in frequent real-world scenarios.

The schedule is thoughtfully arranged to accommodate a global audience, with times listed across multiple time zones so that researchers from different continents can participate without difficulty. This international orientation reflects the collaborative nature of the field, where advances often emerge from cross-institutional and cross-disciplinary partnerships. The workshop serves as a platform for sharing recent results, debating open problems, and identifying promising avenues for future work. By bringing together experts from academia and industry, the program aims to accelerate the translation of research findings into practical systems that can handle the full range of seasonal and environmental conditions.

Beyond the formal talks, the program includes dedicated time for discussion and interaction among participants. These exchanges are designed to spark new ideas and potential collaborations, whether between established research groups or emerging teams entering the field. The workshop organizers have structured the day to balance dense technical content with opportunities for reflection and networking. Participants are encouraged to consider how their own work fits into the broader landscape of robust vision, and how insights from one application domain might transfer to others facing similar challenges under adverse conditions.

The Vision for All Seasons workshop program reflects a growing recognition within the computer vision community that robustness to weather and lighting is not an optional enhancement but a core requirement for many deployed systems. As autonomous vehicles, surveillance networks, and other outdoor applications become more widespread, the ability to perceive reliably under challenging conditions becomes increasingly critical. The talks and discussions scheduled for this edition of the workshop are intended to consolidate recent progress, surface remaining obstacles, and chart a course toward algorithms that maintain high performance regardless of the season or the weather. The organizers hope the program inspires continued research and practical innovation in this important direction.