This work extends the row-column sampling approach to tensors, and introduces a clustering metric that minimizes temporal flicker ( Project). Tensor clustering extends this idea to renderĪnimations including deforming characters. Using them to cluster lights globally, for a fast Pixel-light interactions this matrix can be efficientlyĪpproximated by evaluating a very small set of pixels, and Rendering as the evaluation of a very large matrix of Matrix row-column sampling (SIG '07) treats Scenes with high geometric and illumination complexity, andĪrbitrary shading models, is useful for applications such asĬinematic lighting design. Previewing still images and animations of Scalable previewing for cinematic rendering Lightcuts Project, Subsurface Lightcuts Project, Lightcuts Project). By unifying complex illumination into one framework we achieve high scalability and accurate imagery. Unified, scalable rendering framework to efficiently renderĬomplex scenes with such effects. Multidimensional lightcuts (SIG '06) and lightcuts (SIG '05) present a Participating media, global illumination, and depth-of-field, isĬhallenging. Illumination and rendering effects such as motion blur, Rendering high complexity scenes including complex Visual Equivalence (SIG '07) aims at a new standard of image fidelity that captures what is important in preserving the appearance of objects in a scene Standard image fidelity qualities are limiting and do not necessarily capture what is visually important to a graphics practitioner. We study the perception of aggregates toĭerive metrics for scene simplification (SIG '08) Understanding how we perceive complex geometric aggregates isĪn open problem. We have worked on multiple projects in this area. Understanding and exploiting how humans perceive complex scenes is an important area in graphics. Rendering and modeling complex scenes is challenging. Kavita Bala's Projects Perception of Complex Scenes I captured the interference patterns formed by thin oil films/slicks Satyan Coorg worked on creating the models Rendering model to capture the fine lighting effects of stalactitesĪnd stalagmites. Reconstructing Translucent Objects using Differentiable Renderingĭiscovering Underground Maps from Fashion Zero-shot Learning Using Multimodal Descriptions NeurIPS 22, Datasets and Benchmarks Track James and Mary Tien Excellence in Teaching Award, College of Engineering, 2006Ĭhange-Aware Sampling and Contrastive Learning for Satellite ImagesĬhange Event Dataset for Discovery from Spatio-temporal Remote Sensing Imagery.James and Mary Tien Excellence in Teaching Award, College of Engineering, 2009.Best Paper Award, Computational Aesthetics, 2014.Fiona Ip Li '78 and Donald Li '75 Excellence in Teaching Award, College of Engineering, 2015.ACM SIGGRAPH Computer Graphics Achievement Award, 2020.IIT Bombay Distinguished Alumnus Award, 2021.BTech, CS & Eng., Indian Institute of Technology, Bombay.SM, EECS, Massachusetts Institute of Technology.PhD, EECS, Massachusetts Institute of Technology.Perception: translucency perception material and lighting perception.Rendering: realistic, physically-based rendering scalable rendering.Modeling: material and shape acquisition fabric modeling material representation and editing.Recognition: material recognition, visual search and detection.My research interests span computer vision, computer graphics, and human perception, including:
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |