no code implementations • 6 Feb 2024 • Youngsuk Kim, Hyuk-Jae Lee, Chae Eun Rhee
This paper introduces HEAM, a heterogeneous memory architecture that integrates 3D-stacked DRAM with DIMM to accelerate recommendation systems in which compositional embedding is utilized-a technique aimed at reducing the size of embedding tables.
no code implementations • ICCV 2023 • IlWi Yun, Chanyong Shin, Hyunku Lee, Hyuk-Jae Lee, Chae Eun Rhee
However, to apply local attention successfully for EIs, a specific strategy, which addresses distorted equirectangular geometry and limited receptive field simultaneously, is required.
1 code implementation • 22 Sep 2021 • IlWi Yun, Hyuk-Jae Lee, Chae Eun Rhee
Due to difficulties in acquiring ground truth depth of equirectangular (360) images, the quality and quantity of equirectangular depth data today is insufficient to represent the various scenes in the world.
Ranked #3 on Depth Estimation on Stanford2D3D Panoramic
1 code implementation • ICCV 2021 • Jiwoong Choi, Ismail Elezi, Hyuk-Jae Lee, Clement Farabet, Jose M. Alvarez
Most of these methods are based on multiple models or are straightforward extensions of classification methods, hence estimate an image's informativeness using only the classification head.
no code implementations • 15 Feb 2021 • Heesu Kim, Hanmin Park, Taehyun Kim, Kwanheum Cho, Eojin Lee, Soojung Ryu, Hyuk-Jae Lee, Kiyoung Choi, Jinho Lee
In this paper, we present GradPIM, a processing-in-memory architecture which accelerates parameter updates of deep neural networks training.
no code implementations • 1 Jan 2021 • Jiwoong Choi, Ismail Elezi, Hyuk-Jae Lee, Clement Farabet, Jose M. Alvarez
For active learning, we propose a scoring function that aggregates uncertainties from both the classification and the localization outputs of the network.
no code implementations • 3 Sep 2020 • Duy Thanh Nguyen, Hyun Kim, Hyuk-Jae Lee
The proposed design employs two layer-specific optimizations: layer-specific mixed data flow and layer-specific mixed precision.
4 code implementations • ICCV 2019 • Jiwoong Choi, Dayoung Chun, Hyun Kim, Hyuk-Jae Lee
Therefore, a detection algorithm that can cope with mislocalizations is required in autonomous driving applications.