activity
20182022
most citedD2-Net: A Trainable CNN for Joint Detection and Description of Local Features

74 citations · 74 across the 3 of their papers we have counts for

collaborators
Showing cs.CVShow all

7 papers · 1 filter

cs.CV2022

F2SD: A dataset for end-to-end group detection algorithms

Giang Hoang, Tuan Nguyen Dinh, Tung Cao Hoang +6

The lack of large-scale datasets has been impeding the advance of deep learning approaches to the problem of F-formation detection. Moreover, most research works on this problem re…

cs.CV2020

3D Pipe Network Reconstruction Based on Structure from Motion with Incremental Conic Shape Detection and Cylindrical Constraint

Sho kagami, Hajime Taira, Naoyuki Miyashita +2

Pipe inspection is a critical task for many industries and infrastructure of a city. The 3D information of a pipe can be used for revealing the deformation of the pipe surface and…

cs.CV2019

Is This The Right Place? Geometric-Semantic Pose Verification for Indoor Visual Localization

Hajime Taira, Ignacio Rocco, Jiri Sedlar +5

Visual localization in large and complex indoor scenes, dominated by weakly textured rooms and repeating geometric patterns, is a challenging problem with high practical relevance…

cs.CV201974 cited

D2-Net: A Trainable CNN for Joint Detection and Description of Local Features

Mihai Dusmanu, Ignacio Rocco, Tomas Pajdla +4

In this work we address the problem of finding reliable pixel-level correspondences under difficult imaging conditions. We propose an approach where a single convolutional neural n…

cs.CV2018

Neighbourhood Consensus Networks

Ignacio Rocco, Mircea Cimpoi, Relja Arandjelović +3

We address the problem of finding reliable dense correspondences between a pair of images. This is a challenging task due to strong appearance differences between the corresponding…

cs.CV2018

Structure-from-Motion using Dense CNN Features with Keypoint Relocalization

Aji Resindra Widya, Akihiko Torii, Masatoshi Okutomi

Structure from Motion (SfM) using imagery that involves extreme appearance changes is yet a challenging task due to a loss of feature repeatability. Using feature correspondences o…