Vision and AI Lab, IISc

@iiscbangalore "AdaptiveSplat: Texture Aware Controllable 3D Gaussian Allocation for Feed-Forward Reconstruction" We provide a novel approach to explicitly control the number of 3D Gaussians in feed forward models.

@iiscbangalore "Towards Alias-Free 4D Gaussian Representations with Motion-Aware Filtering" We propose a motion-aware anti-aliasing filter to mitigate aliasing in dynamic representations. By adapting the low-pass filter based on scene motion, our approach improves rendering quality at high res

Thrilled to share that the Vision and AI Lab (@val_iisc), @iiscbangalore, has 3 papers accepted at #ECCV2026! Huge congratulations to all the authors and collaborators for their hard work and dedication! 🧵👇

📣Thrilled that "Rethinking Dataset Distillation: Hard Truths about Soft Labels" is a Best Paper Finalist @CVPR (Top 15 out of 16,000+ submissions)! 🏆 Congrats to the authors! 🥳🙌🏻 Join us today: 🎙️ Oral: 10:15 @ Mile High Ballroom 1A-2A 📍Poster 18 @ ExHall A

Objects in Generated Videos Are Slower Than They Appear: They Follow Sub-Earth Gravity and Don’t Know Galileo’s Principle...for now (Findings) — Varun Varma Thozhiyoor, Shivam Tripathi, R. Venkatesh Babu, and Anand Bhattad. arXiv: arxiv.org/html/2512.0201…

𝘛𝘶𝘳𝘣𝘰-𝘎𝘚: 𝘈𝘤𝘤𝘦𝘭𝘦𝘳𝘢𝘵𝘪𝘯𝘨 𝟥𝘋 𝘎𝘢𝘶𝘴𝘴𝘪𝘢𝘯 𝘍𝘪𝘵𝘵𝘪𝘯𝘨 𝘧𝘰𝘳 𝘏𝘪𝘨𝘩-𝘙𝘦𝘴𝘰𝘭𝘶𝘵𝘪𝘰𝘯 𝘙𝘢𝘥𝘪𝘢𝘯𝘤𝘦 𝘍𝘪𝘦𝘭𝘥𝘴 (main)— Ankit Dhiman, Tao Lu, Srinath R, Emre Arslan, Angela Xing, Yuanbo Xiangli, R. Venkatesh Babu, and Srinath Sridhar. #3DGS

Excited to share that 5 papers (4 main & 1 findings) from the @val_iisc @iiscbangalore have been accepted at @CVPR 2026! 🎉 Our team continues to push boundaries in GenAI, 3D vision, and efficient learning. Check out our accepted works below! 👇🧵 #CVPR2026 #GenAI #IISc

Huge congratulations to Dr. 𝐇𝐚𝐫𝐬𝐡 on receiving the 2025 𝐈𝐊𝐃𝐃 𝐃𝐨𝐜𝐭𝐨𝐫𝐚𝐥 𝐃𝐢𝐬𝐬𝐞𝐫𝐭𝐚𝐭𝐢𝐨𝐧 𝐀𝐰𝐚𝐫𝐝! 🏆 We are incredibly proud to celebrate a 𝐡𝐚𝐭-𝐭𝐫𝐢𝐜𝐤 of successes for VAL:🔹2025: Harsh (Winner)🔹2024: Sravanti (Winner)🔹2023: Jogendra (Runner-Up)

Delighted to share that Prof. R. Venkatesh Babu was inducted as a Fellow of the Indian National Academy of Engineering (INAE) today! 🎖️ This honor recognizes his immense contributions to Computer Vision & AI. A proud moment for VAL, IISc! 🚀 #INAE #Fellow #AI #IISc #CV #ML

[1/3] Two balls, a basketball and a tennis ball, are dropped from the same height, side by side. Assuming no air resistance, which hits the ground first? The answer is neither: they strike the ground simultaneously. Galileo demonstrated this principle over 400 years ago, and it remains one of the most fundamental predictions of Newtonian mechanics. Continuing our theme of what generative models know (and don't) about the physical world -- and building on our #CVPR2024 work exposing failures in projective geometry -- we now revisit Galileo’s equivalence principle. Super excited to share our latest work!

“Make it red.” “No! More red!” “Ughh… slightly less red.” “Perfect!” ♥️ 🎚️Kontinuous Kontext adds slider-based control over edit strength to instruction-based image editing, enabling smooth, continuous transformations!

Prof. Venkatesh delivered a keynote at DAGM German Conference on Pattern Recognition, Freiburg (@gcpr_by_dagm) on “Towards Fair and Controllable Diffusion Models.” sharing the recent works from @val_iisc 🚀 #AI #DiffusionModels #FairAI

VAL reunion at @CVPR📍 A heartwarming moment as current and former members of the Vision and AI Lab (VAL) came together at the conference!—Amazing to see how the VAL family continues to grow—grateful for the bonds, memories, and milestones shared. 💙 @iiscbangalore

Presenting the following papers today @CVPR 1.MirrorVerse: Pushing Diffusion Models to Realistically Reflect the World (Poster - 56) Project:lnkd.in/dWgZcmxh 2.Composing Parts for Expressive Object Generation (Poster - 244) Paper:lnkd.in/dhuSm-AT #CVPR2025 #GenAI

We will be presenting our paper "MonoPlace3D: Learning 3D-Aware Object Placement for 3D Monocular Detection" @CVPR today. 📍Location - Friday, 13th, 16:00 - 18:00 | Poster - 110 📷 Come by the poster to learn more about our work! Project Page: tinyurl.com/bdhc2d6f #CVPR2025

We will be presenting our paper Compass Control: Multi-Object Orientation Control for Text-to-Image Generation 📍Location - Friday, 13th, 10:30 - 12:30 | Poster - 252 💬 Come by the poster to chat more about our work! Project Page - tinyurl.com/u2zhsfan #CVPR2025 #GenAI