Search Results for author: Supasorn Suwajanakorn

Found 18 papers, 9 papers with code

Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoning

1 code implementation NeurIPS 2018 Supasorn Suwajanakorn, Noah Snavely, Jonathan Tompson, Mohammad Norouzi

We demonstrate this framework on 3D pose estimation by proposing a differentiable objective that seeks the optimal set of keypoints for recovering the relative pose between two views of an object.

3D Pose Estimation

Diffusion Autoencoders: Toward a Meaningful and Decodable Representation

2 code implementations CVPR 2022 Konpat Preechakul, Nattanat Chatthee, Suttisak Wizadwongsa, Supasorn Suwajanakorn

Our key idea is to use a learnable encoder for discovering the high-level semantics, and a DPM as the decoder for modeling the remaining stochastic variations.

Attribute Denoising +2

NeX: Real-time View Synthesis with Neural Basis Expansion

1 code implementation CVPR 2021 Suttisak Wizadwongsa, Pakkapon Phongthawee, Jiraphon Yenphraphai, Supasorn Suwajanakorn

We present NeX, a new approach to novel view synthesis based on enhancements of multiplane image (MPI) that can reproduce next-level view-dependent effects -- in real time.

Novel View Synthesis

DiffusionLight: Light Probes for Free by Painting a Chrome Ball

1 code implementation14 Dec 2023 Pakkapon Phongthawee, Worameth Chinchuthakun, Nontaphat Sinsunthithet, Amit Raj, Varun Jampani, Pramook Khungurn, Supasorn Suwajanakorn

To address this problem, we leverage diffusion models trained on billions of standard images to render a chrome ball into the input image.

Lighting Estimation

Accelerating Guided Diffusion Sampling with Splitting Numerical Methods

1 code implementation27 Jan 2023 Suttisak Wizadwongsa, Supasorn Suwajanakorn

Guided diffusion is a technique for conditioning the output of a diffusion model at sampling time without retraining the network for each specific task.

Colorization Super-Resolution +1

Repurposing GANs for One-shot Semantic Part Segmentation

1 code implementation CVPR 2021 Nontawat Tritrong, Pitchaporn Rewatbowornwong, Supasorn Suwajanakorn

Our key idea is to leverage a trained GAN to extract pixel-wise representation from the input image and use it as feature vectors for a segmentation network.

Image Generation Representation Learning +1

SleepPoseNet: Multi-View Learning for Sleep Postural Transition Recognition Using UWB

1 code implementation2 May 2020 Maytus Piriyajitakonkij, Patchanon Warin, Payongkit Lakhan, Pitsharponrn Leelaarporn, Theerasarn Pianpanit, Nakorn Kumchaiseemak, Supasorn Suwajanakorn, Nattee Niparnan, Subhas Chandra Mukhopadhyay, Theerawit Wilaiprasitporn

Recognizing movements during sleep is crucial for the monitoring of patients with sleep disorders, and the utilization of ultra-wideband (UWB) radar for the classification of human sleep postures has not been explored widely.

Data Augmentation Human Activity Recognition +3

Zero-guidance Segmentation Using Zero Segment Labels

1 code implementation ICCV 2023 Pitchaporn Rewatbowornwong, Nattanat Chatthee, Ekapol Chuangsuwanich, Supasorn Suwajanakorn

CLIP has enabled new and exciting joint vision-language applications, one of which is open-vocabulary segmentation, which can locate any segment given an arbitrary text query.

Segmentation Semantic Segmentation +1

Diffusion Sampling with Momentum for Mitigating Divergence Artifacts

1 code implementation20 Jul 2023 Suttisak Wizadwongsa, Worameth Chinchuthakun, Pramook Khungurn, Amit Raj, Supasorn Suwajanakorn

The first technique involves the incorporation of Heavy Ball (HB) momentum, a well-known technique for improving optimization, into existing diffusion numerical methods to expand their stability regions.

Text-to-Image Generation

What Makes Kevin Spacey Look Like Kevin Spacey

no code implementations2 Jun 2015 Supasorn Suwajanakorn, Ira Kemelmacher-Shlizerman, Steve Seitz

We reconstruct a controllable model of a person from a large photo collection that captures his or her {\em persona}, i. e., physical appearance and behavior.

3D Face Reconstruction

Illumination-Aware Age Progression

no code implementations CVPR 2014 Ira Kemelmacher-Shlizerman, Supasorn Suwajanakorn, Steven M. Seitz

We present an approach that takes a single photograph of a child as input and automatically produces a series of age-progressed outputs between 1 and 80 years of age, accounting for pose, expression, and illumination.

Depth From Focus With Your Mobile Phone

no code implementations CVPR 2015 Supasorn Suwajanakorn, Carlos Hernandez, Steven M. Seitz

While prior depth from focus and defocus techniques operated on laboratory scenes, we introduce the first depth from focus (DfF) method capable of handling images from mobile phones and other hand-held cameras.

What Makes Tom Hanks Look Like Tom Hanks

no code implementations ICCV 2015 Supasorn Suwajanakorn, Steven M. Seitz, Ira Kemelmacher-Shlizerman

We reconstruct a controllable model of a person from a large photo collection that captures his or her persona, i. e., physical appearance and behavior.

3D Face Reconstruction

StyleGAN Salon: Multi-View Latent Optimization for Pose-Invariant Hairstyle Transfer

no code implementations CVPR 2023 Sasikarn Khwanmuang, Pakkapon Phongthawee, Patsorn Sangkloy, Supasorn Suwajanakorn

However, there remains a challenge in controlling the hallucinations to accurately transfer hairstyle and preserve the face shape and identity of the input.

Learning Geometric-Aware Properties in 2D Representation Using Lightweight CAD Models, or Zero Real 3D Pairs

no code implementations CVPR 2023 Pattaramanee Arsomngern, Sarana Nutanong, Supasorn Suwajanakorn

We also achieve comparable results to SOTA methods trained on scene scans on four tasks in NYUv2, SUNRGB-D, indoor ADE20k, and indoor/outdoor COCO, despite using lightweight CAD models or pseudo data.

Scene Understanding

Optimizing Diffusion Noise Can Serve As Universal Motion Priors

no code implementations19 Dec 2023 Korrawe Karunratanakul, Konpat Preechakul, Emre Aksan, Thabo Beeler, Supasorn Suwajanakorn, Siyu Tang

We propose Diffusion Noise Optimization (DNO), a new method that effectively leverages existing motion diffusion models as motion priors for a wide range of motion-related tasks.

Denoising

Cannot find the paper you are looking for? You can Submit a new open access paper.