Monitoring the World Through an Imperfect Lens (MONTI) Workshop Published March 17, 2026 By Air Force Capt. Amelia Leonard CAMBRIDGE, Mass. -- The Computer Vision and Pattern Recognition team is seeking short paper submissions on applying machine learning to Earth and environmental science monitoring, particularly focused on multimodal learning under imperfect conditions for the upcoming workshop on Monitoring the World Through an Imperfect Lens (MONTI) scheduled to be held June 3-4, 2026 in Denver, Colo. The goal of this workshop is to gather a wide audience of researchers in academia, industry, and related fields to address real-world constraints in multimodal remote sensing. While many recent multimodal remote sensing publications have focused on adapting computer vision algorithms to satellite imagery, fewer have tackled the unique challenges intrinsic to the remote sensing domain, such as irregular data collection intervals, disparities in modality and resolution, and non-ideal monitoring environments. Modern deep learning approaches show promising results in meteorological applications like precipitation snowcasting, synthetic radar generation, front detection, and several others. The DAF-MIT AI Accelerator is working to rapidly develop new approaches to these challenges. While multimodal remote sensing has the potential to deliver comprehensive Earth observation by combining complementary sensor capabilities there are fundamental challenges prevent this potential from being realized. Computer vision has made remarkable progress in multimodal learning with aligned, simultaneously-collected data (e.g., RGB-D cameras), but remote sensing operates under far more challenging constraints. Satellites collect data asynchronously, at different resolutions, and through fundamentally different imaging physics. Synthetic aperture radar (SAR) actively transmits microwave pulses whereas electro-optical (EO) sensors passively capture reflected sunlight. These disparities create a critical gap between theoretical multimodal methods and practical Earth observation systems. The core challenge in not in sensor alignment or co-registration, but in learning meaningful representations across modalities that differ in their fundamental observational properties. When monitoring dynamic Earth processes, we rarely have the luxury of complete, synchronized observations. Instead, we must extract insights from whatever data is available, such as pre-event optical imagery paired with post-event SAR, high-resolution commercial imagery combined with frequent but coarse images from public satellites, or clear-sky observations from weeks apart bracketing a critical cloudy period. Topics of the upcoming workshop will include, but are not limited to: Multimodal fusion combining EO, SAR, LiDAR, and other sensors Heterogeneous change detection across different modalities Temporal analysis for event monitoring Domain adaptation and cross-sensor generalization Self-/unsupervised learning with limited data Foundation models for remote sensing and Earth observation Uncertainty quantification Real-time multimodal satellite processing Infrastructure monitoring and hazard prediction using incomplete data Multimodal remote sensing analysis Change detection and multi-temporal analysis Geographic Information Science Multimodal generative modeling Multimodal representation learning Speakers include: Ritwik Gupta, Assistant Professor, University of Maryland Johannes Jakubik, Research Scientist, IBM Hannah Kerner, Assistant Professor, Arizona State University Nico Lang, Assistant Professor, University of Copenhagen Esther Rolf, Assistant Professor, University of Colorado Paper Submission Guidelines: Submission must be no more than eight pages (excluding references) on the aforementioned and related topics. We encourage authors to submit four-page papers. Submitted manuscripts should follow the CVPR 2026 paper template. Accepted papers are not archival and will not be included in the proceedings of CVPR 2026. Submissions will be rejected without review if they contain more than eight pages (excluding references), violate the double-blind policy or violate the dual-submission policy. Paper submissions must contain substantial original contents which have not been previously submitted to any other conference, workshop, or journal. Papers will be peer-reviewed under a double-blind policy and will be submitted online through the Open Review submission website. For more information, please contact the CVPR directly here.