Graphics + Vision Research Seminars occur on Mondays in Gates 114, from 2:55-4:10 p.m.

The Graphics + Vision Research Seminar discusses recent research in the areas of computer graphics and computer vision. The goal is to foster technical discussions and collaboration among the Cornell graphics and vision research community.

Fall Schedule TBD

Stay in the loop.

Join the event email list


Date: September 14, 2026

Speaker: Junho Kim
Title: Finding Analogies Everywhere: From 3D Scenes to Touch to Historical Maps
Abstract: Analogies are ubiquitous. From the now-obsolete J.J. Thomson’s “plum pudding” model of the atom to the “greenhouse gases” warming our planet, analogies allow for conveying complex concepts by borrowing illustrations from simpler, well-known  entities. In this talk, I will introduce recent works that find analogies from visual inputs and 3D scenes, which consequently enable transferring complex concepts such as motion trajectories, scene arrangements, and tactile information. Specifically, I will illustrate how 2D/3D foundation model features can be leveraged to find alignments between different domains, and how the alignments can be further used to smoothly map concepts in one domain (e.g., motion or touch) to another. At the end of my talk, I will share progress in my recent work on understanding historical maps, where we aim to find alignments between historical maps and modern maps to model how cities change over time.

Speaker: Chuanruo Ning
Title: Proxy Policy Steering
Abstract: Generalist robot policies carry broad manipulation priors from large-scale data, but specializing them to a new task remains the deployment bottleneck. This requires eliciting task-specific behavior from limited demonstrations without degrading their broad capabilities. We introduce Proxy Policy Steering (PPS), an inference-time adaptation method that resolves this challenge by training two lightweight proxy policies whose calibrated velocity-space difference steers the frozen base sampler. A reference proxy models the frozen base's behavior on target-task observations, and a task proxy, initialized from the reference, captures how this behavior changes under task supervision. Their difference forms a calibrated velocity-space residual that steers the frozen base sampler at every denoising step. We identify the conditions under which this residual isolates the change induced by task supervision, and validate them empirically. Because the base is never directly modified, its broad capabilities remain available at inference, including behaviors such as recovery from failure that the demonstrations themselves do not exercise. Adaptation requires only forward velocity predictions from the base, making PPS lightweight to train and applicable even without access to the base's parameters. On 8 real-world and 4 simulation manipulation tasks, PPS lifts the state-of-the-art pi 0.5 base policy by 53% absolute success rate on average, with zero-to-one gains on tasks the base never solves, while preserving the base's broad capabilities. PPS outperforms LoRA fine-tuning, from-scratch specialists, residual policies, and prior inference-time steering methods.

Event Archive

Browse past lectures.

Date: November 3, 2025
Speaker: Hasindu Piyumantha Kariyawasam Kanaththage
Title: End-to-end Design of Computational Imaging Systems

Speaker: Bradon Thymes
Title: TBD

 

Date: November 17, 2025
Speaker:Bharath Raj
Title: G3T Up: Gravity Aligned Coordinate Frames Simplify Pointmap Prediction

Speaker:Sam Belliveau
Title:Capture Graph: Toward Distribution-Aware Mobile Data Collection

 

Date: November 24, 2025
Speaker:Tao Tu
Title: TBD

Speaker:Mariia Soroka
Title: TBD

 

Date: December 1, 2025
Speaker:Haian Jin
Title: TBD

Speaker:Chao Feng
Title: TBD 
 

Date: December 8, 2025
Speaker: I-Ting Tsai
Title: TBD

Speaker:Chao Zhang
Title: TBD

Date: January 26, 2026
Speaker: Ayush Shrivastava
Title:Point Prompting: Counterfactual Tracking with Video Diffusion Models

Date: January 26, 2026
Speaker: Chia-Hsiang Kao
Title:Δynamics: Language-Based Representation for Inferring Rigid-Body Dynamics From Videos
 

Date: February 2, 2026
Speaker: Evan Zhang
Title:Emergent Extreme-View Geometry in 3D Foundation Models

Date: February 2, 2026
Speaker: Rundong Luo
Title:Perceiving and Manipulating the Flow of Time
 

Date: February 9, 2026
Speaker: William Yang, Princeton University
Title:Towards Synthesizing More Informative Task-driven Datasets
 

Date: March 2, 2026
ECCV Abstracts
Speaker: Noam Atia
Title: BlendedPC: Blended Point Cloud Diffusion for Localized Text-guided Shape Editing
 

Date: March 9, 2026
Speaker: Xuanchen Lu
Title:Generative Point Tracking and Trajectory Forecasting


Date: March 9, 2026
Speaker: Ishit Mehta
Title:Inverse Rendering of Geometry
 

Date: March 16, 2026
Speaker: Shenlong Wang
Title:May the Force Be with Your Pixels
 

Date: March 23, 2026
Speaker: Blaire Yu
Title: Toward Richer Material Generation via Procedural Data Enhancement
 

Date: March 23, 2026
Speaker: Hao Phung
Title:ProxE: Fine-Grained 3D Shape Editing via Primitive-Based Abstractions
 

Date: April 6, 2026
Speaker: Rajeev Datta
Title: Zero-Shot Concept Bottlenecks: A Reality Check
 

Date: April 13, 2026
Speaker: Junhyeong Cho
Title:SceneAligner: 3D-Grounded Floorplan Localization in the Wild

Date: April 13, 2026
Speaker: Nhan Tran
Title: CineCraft: Unified Shot Planning, Capture, and Post-Processing for Mobile Cinematography
 

Date: April 20, 2026
Speaker: Longxiulin Deng
Title: Performance-Integrated Production for Kinetic Typography & The user as an objective function: Using interaction design to improve throughput in active learning
 

Date: April 27, 2026
NYC Vision Day
 

Date: May 4, 2026
Speaker: Salma Abdel Magid, Princeton University
Title:Auditing the Text-to-Image Feedback Loop