Anand Bhattad
Research Assistant Professor
Toyota Technological Institute at Chicago
Visiting Research Scholar
UC Berkeley
Email /
CV /
Google Scholar
/
Twitter
Top 3 Research Highlights: ppt, pdf
|
|
|
I am a Research Assistant Professor at the Toyota Technological Institute at Chicago (TTIC) located on the University of Chicago campus. I am also a visiting research scholar at UC Berkeley, hosted by Alexei (Alyosha) Efros.
Earlier, I earned my PhD from Computer Vision Group at University of Illinois Urbana-Champaign (UIUC), where my advisor was David Forsyth, and I worked closely with Derek Hoiem, Shenlong Wang, and Yuxiong Wang during my PhD.
Research
My research is centered around the fields of computer vision, computer graphics, computational
photography, and machine learning. I work on developing methods that are grounded in real-world
physics to better understand, model, and recreate what we see around us. At the moment, my focus is
on generative models to explore the kind of knowledge they capture and their limitations. My
goal is to recover inherent scene characteristics these models learn and apply them to
build advanced computer vision systems capable of creating and modeling digital twins.
I am also interested in leveraging generative models for neural rendering and 3D-aware
image editing. This aims to provide easy-to-use tools for interacting with and manipulating visual
content.
Collaborators
My recent and ongoing collaborations include working with
News
-
July 2024: VideoShop has been accepted to ECCV 2024! Congratulations, Xiang Fan!
-
May 2024: Spending some time with Alyosha and the BAIR folks at UC Berkeley.
-
Mar 2024: Teaching a new course: Past
Meets Present: A Tale of Two Visions at TTIC this Spring.
-
Feb 2024: Organzing CV 20/20: A
Retrospective Vision workshop at CVPR 2024.
-
Feb 2024: Congratulations to James Soole, Ayush Sarkar, Asher Mai and Amitabh
Mahapatra on their CVPR 2024 papers!
-
Sept 2023: Congratulations to Oscar Michel on his NeurIPS 2023 paper acceptance!
Sept 2023: Excited to start as a Research Assistant Professor at TTIC!
-
July 2023: One paper accepted by ICCV 2023 on Equivariant Dense Prediction
Models.
-
Jun 2023: Organized Scholars
& Big Models: How Can Academics Adapt? workshop at CVPR 2023.
-
May 2023: Successfully defended my PhD thesis titled "Exploring Knowledge in
Generative Models"!
Recent and Upcoming Talks
- IIIT-Hyderabad, Jan 2024
- TTIC, Oct 2023
- Exactech, Inc., Oct 2023
- Stanford University, Jun 2023
- University of Tübingen, Autonomous Vision Group, May 2023
- UC Berkeley: Vision Seminar, Apr 2023
- NVIDIA Research, Apr 2023
- MIT: Vision and Graphics Seminar, Apr 2023
- CMU: VASC Seminar, Mar 2023
- UW: Vision Seminar, Mar 2023
- UMD: Vision Seminar, Mar 2023
- UCSD: Pixel Cafe Seminar, Feb 2023
- TTIC: Research Talk, Feb 2023
Recent Awards
- Outstanding Reviewer Award, ICCV 2023
- Best Paper Finalist, CVPR 2022
- Outstanding Emergency Reviewer Award, CVPR 2021
|
Preprints
|
|
Xiao Zhang, William Gao, Seemandhar Jain, Michael Maire, David Forsyth,
Anand Bhattad
arXiv, 2024
[arXiv] [project page]
Free albedo extraction from latent features in a relighting-trained model, without seeing albedo like images
|
|
Vaibhav Vavilala, Seemandhar Jain*, Rahul Vasanth*, Anand Bhattad, David Forsyth
arXiv, 2023
arXiv
Blocks2World decomposes 3D scenes into editable primitives and uses a trained model to render
these
into 2D images, providing high control for scene editing.
|
|
Make It So: Steering StyleGAN for Any Image Inversion and Editing
Anand Bhattad,
Viraj Shah
Derek Hoiem,
David A. Forsyth
arXiv, 2023
[arXiv], [project page]
A novel near-perfect GAN Inversion method that preserves editing capabilities, even for
out-of-domain images
|
|
SIRfyN: Single Image Relighting from your Neighbors
David A. Forsyth,
Anand Bhattad,
Pranav Asthana,
Yuani Zhong,
Yuxiong Wang
arXiv, 2022
Technical Report
First scene relighting method that requires no
labeled or paired image data.
|
2024
|
|
Xiang Fan,
Anand Bhattad*, Ranjay Krishna*
ECCV, 2024
[arXiv]
[project page]
Videoshop is a training-free algorithm that allows users to edit videos by modifying only the first frame using familiar image editing tools. It automatically propagates these edits throughout the video, maintaining consistency and allowing for precise, localized semantic changes.
|
|
Ayush Sarkar*, Hanlin Mai*, Amitabh Mahapatra*, Svetlana Lazebnik, David Forsyth, Anand Bhattad
CVPR, 2024
[arXiv]
[project page]
[code]
Generative models are not aware of projective geometry. We show that generated images can be
easily
distinguished by looking at derived projective geometry cues.
|
|
Xiaodan Du,
Nicholas Kolkin, Greg Shakhnarovich, Anand Bhattad
CVPR-W, 2024 (short paper)
arXiv, 2023
[arXiv] [project page] [code]
Intrinsic LoRA: A method to extract intrinsic images from ANY generative model, be it
Autoregressive, GAN, Diffusion!
|
|
StyLitGAN: Image-based Relighting via Latent Control
Anand Bhattad, James Soole,
David A. Forsyth
CVPR, 2024
[paper]
[arXiv]
[project page]
[code]
By imposing known physical facts about images, we can prompt StyleGAN to generate relighted or
resurfaced images without using labeled data.
|
|
Zhi-Hao Lin, Bohan Liu, Yi-Ting Chen, David Forsyth, Jia-Bin Huang, Anand Bhattad, Shenlong Wang
CVPR-W, 2024 (short paper)
arXiv, 2023
[arXiv] [project page]
UrbanIR creates realistic 3D renderings of urban scenes from single videos, allowing for novel
lighting conditions and controllable editing.
|
2023
|
|
Anand Bhattad, Daniel McKee, Derek Hoiem, David Forsyth
NeurIPS, 2023
[arXiv]
StyleGAN has easy accssible internal encoding of intrinsic images as originally defined by Barrow
and Tenenbaum in their influential paper of 1978.
|
|
Oscar Michel, Anand Bhattad, Eli VanderBilt, Ranjay Krishna, Ani
Kembhavi, Tanmay Gupta
NeurIPS, 2023
[arXiv] [project page]
A synthetic dataset and a model that learns to rotate, translate, insert, and remove objects
identified by language in a scene. It can transfer to real-world images.
|
|
Yuani Zhong,
Anand Bhattad,
Yuxiong Wang
David A. Forsyth
ICCV 2023
SOTA normal and depth predictors are not equivariant to image cropping. We propose equivariant
regularization loss to improve equivariance in these models.
|
2022
|
|
Cut-and-Paste Object Insertion by Enabling Deep Image Prior for Reshading
Anand Bhattad,
David A. Forsyth
3DV, 2022
[project page]
Convincing cut-and-paste reshading with consistent image decomposition inferences.
|
|
DIVeR: Real-time and Accurate Neural Radiance Fields with Deterministic
Integration for
Volume Rendering
Liwen Wu,
Jae Yong Lee,
Anand Bhattad,
Yuxiong Wang,
David A. Forsyth
CVPR, 2022 (Best Paper Finalist)
[project page] /
Training Code /
Real-time Code
Improving Real-Time NeRF with Deterministic Integration.
|
2021
|
|
View Generalization for Single Image Textured 3D Models
Anand Bhattad,
Aysegul
Dundar,
Guilin Liu,
Andrew
Tao,
Bryan Catanzaro
CVPR, 2021
[project page]
Consistent textured 3D inferences from a single 2D image.
|
2020
|
|
Unrestricted Adversarial Examples via Semantic Manipulation
Anand Bhattad*,
Min Jin Chong*,
Kaizhao Liang,
Bo Li,
David A. Forsyth
ICLR, 2020
Generating realistic adversarial examples by image re-colorization and texture transfer.
|
|
Improved Style Transfer with Calibrated Metrics
Mao Chuang Yeh*,
Shuai Tang*,
Anand Bhattad,
Chuhang Zou,
David A. Forsyth
WACV, 2020
A novel quantitative evaluation procedure for style transfer methods.
|
2018
|
|
Detecting Anomalous Faces with "No Peeking'' Autoencoders
Anand Bhattad,
Jason
Rock,
David A. Forsyth
CVPR Workshop on Vision
with Biased or
Scarce Data, 2018
A simple unsupervised method for detecting anomalous faces by carefully constructing features
from
"No Peeking" or inpainting autoencoders.
|
|