Seung-Hwan Baek

Seung-Hwan Baek

Auto-generated profile, based on publicly available information

Seung-Hwan Baek

POSTECH, Pohang, Republic of Korea

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 June 2024 in Nature Photonics
Verified
Journal Article
Closed access
PDF file
ECEunsue Choi
Eunsue Choi
GKGyeongtae Kim
Gyeongtae Kim
JYJooyeong Yun
Jooyeong Yun
YJYujin Jeon
Yujin Jeon
Junsuk RhoJunsuk Rho
Junsuk Rho
SBSeung-Hwan Baek
Seung-Hwan Baek

29 November 2022
Verified
Conference Paper
Open access
QZQiang Zhang
Qiang Zhang
SBSeung-Hwan Baek
Seung-Hwan Baek
SRSzymon Rusinkiewicz
Szymon Rusinkiewicz
FHFelix Heide
Felix Heide
We propose a differentiable rendering algorithm for efficient novel view synthesis. By departing from volume-based representations in favor of a learned point representation, we improve on existing methods more than an order of magnitude in memory and runtime, both in training and inference. The method begins with a uniformly-sampled random point cloud and learns per-point position and view-dependent appearance, using a differentiable splat-based renderer to train the model to reproduce a set of input training images with the given pose. Our method is up to 300 × faster than NeRF in both training and inference, with only a marginal sacrifice in quality, while using less than 10 MB of memory for a static scene. For dynamic scenes, our method trains two orders of magnitude faster than STNeRF and renders at a near interactive rate, while maintaining high image quality and temporal coherence even without imposing any temporal-coherency regularizers.

13 November 2022
Verified
Book Chapter
Closed access
PDF file
GKGeonung Kim
Geonung Kim
KKKyoungkook Kang
Kyoungkook Kang
SKSeongtae Kim
Seongtae Kim
HLHwayoon Lee
Hwayoon Lee
SKSehoon Kim
Sehoon Kim
SCSunghyun Cho
Sunghyun Cho

28 February 2023 in ACM Transactions on Graphics
Verified
Journal Article
Open access
SBSeung-Hwan Baek
Seung-Hwan Baek
NWNoah Walsh
Noah Walsh
ICIlya Chugunov
Ilya Chugunov
ZSZheng Shi
Zheng Shi
FHFelix Heide
Felix Heide
Depth sensors have emerged as a cornerstone sensor modality with diverse applications in personal hand-held devices, robotics, scientific imaging, autonomous vehicles, and more. In particular, correlation Time-of-Flight (ToF) sensors have found widespread adoption for meter-scale indoor applications such as object tracking and pose estimation. While they offer high depth resolution at competitive costs, the precision of these indirect ToF sensors is fundamentally limited by their modulation contrast, which is in turn limited by the effects of photo-conversion noise. In contrast, optical interferometric methods can leverage short illumination modulation wavelengths to achieve depth precision three orders of magnitude greater than ToF, but typically find their range is restricted to the sub-centimeter. In this work, we merge concepts from both correlation ToF design and interferometric imaging; a step towards bridging the gap between these methods. We propose a computational ToF imaging method that optically computes the GHz ToF correlation signal in free space before photo-conversion. To acquire a depth map, we scan a scene point-wise and computationally unwrap the collected correlation measurements. Specifically, we repurpose electro-optical modulators used in optical communication for ToF imaging with centimeter-wave signals, and achieve all-optical correlation at 7.15 GHz and 14.32 GHz modulation frequencies. While GHz modulation frequencies increase depth precision, these high modulation rates also pose a technical challenge. They result in dozens of wraps per meter which cannot be estimated robustly by existing phase unwrapping methods. We tackle this problem with a proposed segmentation-inspired phase unwrapping network , which exploits the correlation of adjacent GHz phase measurements to classify regions into their respective wrap counts. We validate this method in simulation and experimentally, and demonstrate precise depth sensing using centimeter wave modulation that is robust to surface texture and ambient light. Compared to existing analog demodulation methods, the proposed system outperforms all of them across all tested scenarios. CCS Concepts: • Computing methodologies ;

1 July 2025 in Nature Communications
Verified
Journal Article
Open access
PDF file
MCMinho Choi
Minho Choi
JXJinlin Xiang
Jinlin Xiang
AWAnna Wirth-Singh
Anna Wirth-Singh
SBSeung-Hwan Baek
Seung-Hwan Baek
ESEli Shlizerman
Eli Shlizerman
AMArka Majumdar
Arka Majumdar
Artificial neural networks have fundamentally transformed the field of computer vision, providing unprecedented performance. However, these neural networks for image processing demand substantial computational resources, often hindering real-time operation. In this work, we demonstrate an optical encoder that can perform convolution simultaneously in three color channels during the image capture, effectively implementing several initial convolutional layers of the network. Such an optical encoding results in ~ 24, 000 × reduction in computational operations, with a state-of-the-art classification accuracy (~73.2%) in free-space optical system. In addition, our analog optical encoder, trained for CIFAR-10 data, can be transferred to the ImageNet subset, High-10, without any modifications, and still exhibits moderate accuracy. The proposed method can decrease total system-level energy more than two orders of magnitude per a single object classification. Our results evidence the potential of hybrid optical/digital computer vision system in which the optical frontend can pre-process an ambient scene to reduce the energy and latency of the whole computer vision system.

29 November 2022
Verified
Conference Paper
Open access
SKSeongtae Kim
Seongtae Kim
KKKyoungkook Kang
Kyoungkook Kang
GKGeonung Kim
Geonung Kim
SBSeung-Hwan Baek
Seung-Hwan Baek
SCSunghyun Cho
Sunghyun Cho
Few-shot domain adaptation to multiple domains aims to learn a complex image distribution across multiple domains from a few training images. A naïve solution here is to train a separate model for each domain using few-shot domain adaptation methods. Unfortunately, this approach mandates linearly-scaled computational resources both in memory and computation time and, more importantly, such separate models cannot exploit the shared knowledge between target domains. In this paper, we propose DynaGAN, a novel few-shot domain-adaptation method for multiple target domains. DynaGAN has an adaptation module, which is a hyper-network that dynamically adapts a pretrained GAN model into the multiple target domains. Hence, we can fully exploit the shared knowledge across target domains and avoid the linearly-scaled computational requirements. As it is still computationally challenging to adapt a large-size GAN model, we design our adaptation module to be lightweight using the rank-1 tensor decomposition. Lastly, we propose a contrastive-adaptation loss suitable for multi-domain few-shot adaptation. We validate the effectiveness of our method through extensive qualitative and quantitative evaluations.