Stanford AI Lab Papers and Talks at CVPR 2021

The Conference on Computer Vision and Pattern Recognition (CVPR) 2021 is being hosted virtually from June 19th – June 25th. We’re excited to share all the work from SAIL that’s being presented, and you’ll find links to papers, videos and blogs below. Feel free to reach out to the contact authors directly to learn more about the work that’s happening at Stanford!

List of Accepted Papers

GeoSim: Realistic Video Simulation via Geometry-Aware Composition for Self-Driving


Authors: Yun Chen*, Frieda Rong*, Shivam Duggal*, Shenlong Wang, Xinchen Yan, Sivabalan Manivasagam, Shangjie Xue, Ersin Yumer, Raquel Urtasun

Contact: chenyuntc@gmail.com

Award nominations: Oral, Best Paper Finalist

Links: Paper | Video | Website

Keywords: computer vision, simulation, image simulation, video simulation, self-driving, autonomous driving, 3d vision, computer graphics, robotics


Greedy hierarchical variational autoencoders for large-scale video prediction


Authors: Bohan Wu, Suraj Nair, Roberto Martin-Martin, Li Fei-Fei*, Chelsea Finn*

Contact: bohanwu@stanford.edu

Keywords: variational autoencoders, video prediction


AGQA: A Benchmark for Compositional Spatio-Temporal Reasoning


Authors: Madeleine Grunde-McLaughlin

Contact: mgrund@sas.upenn.edu

Links: Paper | Video | Website

Keywords: visual question answering, compositionality, computer vision, benchmark


ArtEmis: Affective Language for Visual Art


Authors: Panos Achlioptas, Maks Ovsjanikov, Kilichbek Haydarov, Mohamed Elhoseiny, Leonidas Guibas

Contact: panos@cs.stanford.edu

Award nominations: Oral

Links: Paper | Video | Website

Keywords: affective-computing, wikiart, neural-speakers, emotions


DARCNN: Domain Adaptive Region-based Convolutional Neural Network for Unsupervised Instance Segmentation in Biomedical Images


Authors: Joy Hsu, Wah Chiu, Serena Yeung

Contact: joycj@stanford.edu

Links: Paper | Website

Keywords: unsupervised domain adaptation, instance segmentation


Hierarchical Motion Understanding via Motion Programs

Authors: Sumith Kulal*, Jiayuan Mao*, Alex Aiken, Jiajun Wu

Contact: sumith@cs.stanford.edu

Links: Paper | Video | Website

Keywords: neuro-symbolic, motion, primitives, programs


Home Action Genome: Cooperative Compositional Action Understanding


Authors: Nishant Rai

Contact: nishantr018@gmail.com

Links: Paper | Website

Keywords: multi modal, multi camera view, multi perspective, action recognition, action localization, atomic actions, scene graphs, contrastive learning, audio-visual, large scale dataset


Joint Learning of 3D Shape Retrieval and Deformation


Authors: Mikaela Angelina Uy, Vladimir G. Kim, Minhyuk Sung, Noam Aigerman, Siddhartha Chaudhuri, Leonidas Guibas

Contact: mikacuy@stanford.edu

Links: Paper | Video | Website

Keywords: joint learning, retrieval, deformation


Metadata Normalization


Authors: Mandy Lu, Qingyu Zhao, Jiequan Zhang, Kilian M. Pohl, Li Fei-Fei, Juan Carlos Niebles, Ehsan Adeli

Contact: mlu@cs.stanford.edu

Links: Paper | Website

Keywords: metadata, normalization, bias, deep learning, bias-free feature learning


We look forward to seeing you at CVPR 2021!

Read More