activity
20182025
most citedUR2KiD: Unifying Retrieval, Keypoint Detection, and Keypoint Description without Local Correspondence Supervision

27 citations · 38 across the 5 of their papers we have counts for

collaborators

9 papers

cs.CV2025

Monocular Visual Odometry without Calibration or Test-time Optimization

Vladimir Yugay, Vlardimir Yugay, Duy-Kien Nguyen +3

The most accurate monocular visual odometry systems require known camera intrinsics, refine their estimates with test-time optimization, and recover trajectories only up to an unkn…

cs.CV2024★ 10 cited

An Image is Worth More Than 16x16 Patches: Exploring Transformers on Individual Pixels

Duy-Kien Nguyen, Mahmoud Assran, Unnat Jain +3

This work does not introduce a new method. Instead, we present an interesting finding that questions the necessity of the inductive bias of locality in modern computer vision archi…

cs.CV2023

SimPLR: A Simple and Plain Transformer for Efficient Object Detection and Segmentation

Duy-Kien Nguyen, Martin R. Oswald, Cees G. M. Snoek

The ability to detect objects in images at varying scales has played a pivotal role in the design of modern object detectors. Despite considerable progress in removing hand-crafted…

cs.CV2023★ 1 cited

R-MAE: Regions Meet Masked Autoencoders

Duy-Kien Nguyen, Vaibhav Aggarwal, Yanghao Li +4

In this work, we explore regions as a potential visual analogue of words for self-supervised image representation learning. Inspired by Masked Autoencoding (MAE), a generative pre-…

cs.CV2021

BoxeR: Box-Attention for 2D and 3D Transformers

Duy-Kien Nguyen, Jihong Ju, Olaf Booij +2

In this paper, we propose a simple attention mechanism, we call box-attention. It enables spatial interaction between grid features, as sampled from boxes of interest, and improves…

cs.CV2020

MoVie: Revisiting Modulated Convolutions for Visual Counting and Beyond

Duy-Kien Nguyen, Vedanuj Goswami, Xinlei Chen

This paper focuses on visual counting, which aims to predict the number of occurrences given a natural image and a query (e.g. a question or a category). Unlike most prior works th…