Search Results for author: Gim Hee Lee

Found 95 papers, 57 papers with code

Relative Pose Estimation of Calibrated Cameras with Known SE(3) Invariants

1 code implementation ECCV 2020 Bo Li, Evgeniy Martyushev, Gim Hee Lee

In this paper, we present a complete comprehensive study of the relative pose estimation problem for a calibrated camera constrained by known $\mathrm{SE}(3)$ invariant, which involves 5 minimal problems in total.

Pose Estimation Translation

GeoGaussian: Geometry-aware Gaussian Splatting for Scene Rendering

no code implementations17 Mar 2024 Yanyan Li, Chenyu Lyu, Yan Di, Guangyao Zhai, Gim Hee Lee, Federico Tombari

During the Gaussian Splatting optimization process, the scene's geometry can gradually deteriorate if its structure is not deliberately preserved, especially in non-textured regions such as walls, ceilings, and furniture surfaces.

Novel View Synthesis

URS-NeRF: Unordered Rolling Shutter Bundle Adjustment for Neural Radiance Fields

no code implementations15 Mar 2024 Bo Xu, Ziao Liu, Mengqi Guo, Jiancheng Li, Gim Hee Lee

We propose a novel rolling shutter bundle adjustment method for neural radiance fields (NeRF), which utilizes the unordered rolling shutter (RS) images to obtain the implicit 3D representation.

NeuSG: Neural Implicit Surface Reconstruction with 3D Gaussian Splatting Guidance

no code implementations1 Dec 2023 Hanlin Chen, Chen Li, Gim Hee Lee

In this work, we propose a neural implicit surface reconstruction pipeline with guidance from 3D Gaussian Splatting to recover highly detailed surfaces.

3D Reconstruction Multi-View 3D Reconstruction +1

Multi-Scale 3D Gaussian Splatting for Anti-Aliased Rendering

no code implementations28 Nov 2023 Zhiwen Yan, Weng Fei Low, Yu Chen, Gim Hee Lee

3D Gaussians have recently emerged as a highly efficient representation for 3D reconstruction and rendering.

3D Reconstruction

SCALAR-NeRF: SCAlable LARge-scale Neural Radiance Fields for Scene Reconstruction

no code implementations28 Nov 2023 Yu Chen, Gim Hee Lee

We enhance the overlapping regions across different blocks by scaling up the bounding boxes of each local block.

Animate124: Animating One Image to 4D Dynamic Scene

no code implementations24 Nov 2023 Yuyang Zhao, Zhiwen Yan, Enze Xie, Lanqing Hong, Zhenguo Li, Gim Hee Lee

We introduce Animate124 (Animate-one-image-to-4D), the first work to animate a single in-the-wild image into 3D video through textual motion descriptions, an underexplored problem with significant applications.

Rethink Cross-Modal Fusion in Weakly-Supervised Audio-Visual Video Parsing

no code implementations14 Nov 2023 Yating Xu, Conghui Hu, Gim Hee Lee

Existing works on weakly-supervised audio-visual video parsing adopt hybrid attention network (HAN) as the multi-modal embedding to capture the cross-modal context.

Generalized Few-Shot Point Cloud Segmentation Via Geometric Words

1 code implementation ICCV 2023 Yating Xu, Conghui Hu, Na Zhao, Gim Hee Lee

Existing fully-supervised point cloud segmentation methods suffer in the dynamic testing environment with emerging new classes.

Point Cloud Segmentation Segmentation

Towards Robust Few-shot Point Cloud Semantic Segmentation

1 code implementation20 Sep 2023 Yating Xu, Na Zhao, Gim Hee Lee

Few-shot point cloud semantic segmentation aims to train a model to quickly adapt to new unseen classes with only a handful of support set samples.

Point Cloud Segmentation Representation Learning +1

Robust e-NeRF: NeRF from Sparse & Noisy Events under Non-Uniform Motion

1 code implementation ICCV 2023 Weng Fei Low, Gim Hee Lee

Such promise and potential of event cameras and NeRF inspired recent works to investigate on the reconstruction of NeRF from moving event cameras.

3D Scene Reconstruction Novel View Synthesis

GHuNeRF: Generalizable Human NeRF from a Monocular Video

1 code implementation31 Aug 2023 Chen Li, Jiahao Lin, Gim Hee Lee

In view of these limitations, we propose GHuNeRF to learn a generalizable human NeRF model from a monocular video of the human performer.

GeT: Generative Target Structure Debiasing for Domain Adaptation

no code implementations ICCV 2023 Can Zhang, Gim Hee Lee

Despite the competitive performance, these pseudo labeling methods rely heavily on the source domain to generate pseudo labels for the target domain and therefore still suffer considerably from source data bias.

Domain Adaptation Pseudo Label

DReg-NeRF: Deep Registration for Neural Radiance Fields

1 code implementation ICCV 2023 Yu Chen, Gim Hee Lee

Although Neural Radiance Fields (NeRF) is popular in the computer vision community recently, registering multiple NeRFs has yet to gain much attention.

Novel View Synthesis Point Cloud Registration +1

Weakly-supervised 3D Pose Transfer with Keypoints

1 code implementation ICCV 2023 Jinnan Chen, Chen Li, Gim Hee Lee

The main challenges of 3D pose transfer are: 1) Lack of paired training data with different characters performing the same pose; 2) Disentangling pose and shape information from the target mesh; 3) Difficulty in applying to meshes with different topologies.

Pose Transfer

OD-NeRF: Efficient Training of On-the-Fly Dynamic Neural Radiance Fields

no code implementations24 May 2023 Zhiwen Yan, Chen Li, Gim Hee Lee

Dynamic neural radiance fields (dynamic NeRFs) have demonstrated impressive results in novel view synthesis on 3D dynamic scenes.

Novel View Synthesis

Boosting Long-tailed Object Detection via Step-wise Learning on Smooth-tail Data

no code implementations ICCV 2023 Na Dong, Yongqiang Zhang, Mingli Ding, Gim Hee Lee

Real-world data tends to follow a long-tailed distribution, where the class imbalance results in dominance of the head classes during training.

Long-tailed Object Detection object-detection +1

ConsistentNeRF: Enhancing Neural Radiance Fields with 3D Consistency for Sparse View Synthesis

1 code implementation18 May 2023 Shoukang Hu, Kaichen Zhou, Kaiyu Li, Longhui Yu, Lanqing Hong, Tianyang Hu, Zhenguo Li, Gim Hee Lee, Ziwei Liu

In this paper, we propose ConsistentNeRF, a method that leverages depth information to regularize both multi-view and single-view 3D consistency among pixels.

3D Reconstruction SSIM

Make-A-Protagonist: Generic Video Editing with An Ensemble of Experts

no code implementations15 May 2023 Yuyang Zhao, Enze Xie, Lanqing Hong, Zhenguo Li, Gim Hee Lee

The text-driven image and video diffusion models have achieved unprecedented success in generating realistic and diverse content.

Denoising Video Editing +1

ScarceNet: Animal Pose Estimation with Scarce Annotations

1 code implementation CVPR 2023 Chen Li, Gim Hee Lee

To this end, we propose the ScarceNet, a pseudo label-based approach to generate artificial labels for the unlabeled images.

Animal Pose Estimation Domain Adaptation +1

DBARF: Deep Bundle-Adjusting Generalizable Neural Radiance Fields

no code implementations CVPR 2023 Yu Chen, Gim Hee Lee

Recent works such as BARF and GARF can bundle adjust camera poses with neural radiance fields (NeRF) which is based on coordinate-MLPs.

NeRF-DS: Neural Radiance Fields for Dynamic Specular Objects

1 code implementation CVPR 2023 Zhiwen Yan, Chen Li, Gim Hee Lee

We evaluate our model based on the novel view synthesis quality with a self-collected dataset of different moving specular objects in realistic environments.

Novel View Synthesis

ContraNeRF: Generalizable Neural Radiance Fields for Synthetic-to-real Novel View Synthesis via Contrastive Learning

no code implementations CVPR 2023 Hao Yang, Lanqing Hong, Aoxue Li, Tianyang Hu, Zhenguo Li, Gim Hee Lee, LiWei Wang

In this work, we first investigate the effects of synthetic data in synthetic-to-real novel view synthesis and surprisingly observe that models trained with synthetic data tend to produce sharper but less accurate volume densities.

Contrastive Learning Generalizable Novel View Synthesis +2

AdaSfM: From Coarse Global to Fine Incremental Adaptive Structure from Motion

no code implementations28 Jan 2023 Yu Chen, Zihao Yu, Shu Song, Tianning Yu, Jianming Li, Gim Hee Lee

Despite the impressive results achieved by many existing Structure from Motion (SfM) approaches, there is still a need to improve the robustness, accuracy, and efficiency on large-scale scenes with many outlier matches and sparse view graphs.

Unsupervised Feature Representation Learning for Domain-generalized Cross-domain Image Retrieval

1 code implementation ICCV 2023 Conghui Hu, Can Zhang, Gim Hee Lee

This limitation motivates us to present the first attempt at domain-generalized unsupervised cross-domain image retrieval (DG-UCDIR) aiming at facilitating image retrieval between any two unseen domains in an unsupervised way.

Contrastive Learning Image Retrieval +2

Style-Hallucinated Dual Consistency Learning: A Unified Framework for Visual Domain Generalization

1 code implementation18 Dec 2022 Yuyang Zhao, Zhun Zhong, Na Zhao, Nicu Sebe, Gim Hee Lee

Furthermore, we present a novel style hallucination module (SHM) to generate style-diversified samples that are essential to consistency learning.

Domain Generalization Hallucination +4

Motion and Context-Aware Audio-Visual Conditioned Video Prediction

no code implementations9 Dec 2022 Yating Xu, Conghui Hu, Gim Hee Lee

The existing state-of-the-art method for audio-visual conditioned video prediction uses the latent codes of the audio-visual frames from a multimodal stochastic network and a frame encoder to predict the next visual frame.

Motion Estimation Optical Flow Estimation +1

Synthetic-to-Real Domain Generalized Semantic Segmentation for 3D Indoor Point Clouds

no code implementations9 Dec 2022 Yuyang Zhao, Na Zhao, Gim Hee Lee

In addition, we augment the point patterns of the source data and introduce non-parametric multi-prototypes to ameliorate the intra-class variance enlarged by the augmented point patterns.

Domain Generalization Semantic Segmentation

Open World DETR: Transformer based Open World Object Detection

no code implementations6 Dec 2022 Na Dong, Yongqiang Zhang, Mingli Ding, Gim Hee Lee

Open world object detection aims at detecting objects that are absent in the object classes of the training data as unknown objects without explicit supervision.

Knowledge Distillation Object +2

Minimal Neural Atlas: Parameterizing Complex Surfaces with Minimal Charts and Distortion

1 code implementation29 Jul 2022 Weng Fei Low, Gim Hee Lee

Thus, our representation can learn a minimal atlas of 3 charts with distortion-minimal parameterization for surfaces of arbitrary topology, including closed and open surfaces with arbitrary connected components.

Surface Reconstruction

Rethinking IoU-based Optimization for Single-stage 3D Object Detection

1 code implementation19 Jul 2022 Hualian Sheng, Sijia Cai, Na Zhao, Bing Deng, Jianqiang Huang, Xian-Sheng Hua, Min-Jian Zhao, Gim Hee Lee

Since Intersection-over-Union (IoU) based optimization maintains the consistency of the final IoU prediction metric and losses, it has been widely used in both regression and classification branches of single-stage 2D object detectors.

3D Object Detection Object +1

Adversarial Style Augmentation for Domain Generalized Urban-Scene Segmentation

1 code implementation11 Jul 2022 Zhun Zhong, Yuyang Zhao, Gim Hee Lee, Nicu Sebe

Experiments on two synthetic-to-real semantic segmentation benchmarks demonstrate that AdvStyle can significantly improve the model performance on unseen real domains and show that we can achieve the state of the art.

Domain Generalization Image Classification +1

CA-UDA: Class-Aware Unsupervised Domain Adaptation with Optimal Assignment and Pseudo-Label Refinement

no code implementations26 May 2022 Can Zhang, Gim Hee Lee

However, source domain bias that deteriorates the pseudo-labels can still exist since the shared network of the source and target domains are typically used for the pseudo-label selections.

Image Classification Missing Labels +2

VRAG: Region Attention Graphs for Content-Based Video Retrieval

no code implementations18 May 2022 Kennard Ng, Ser-Nam Lim, Gim Hee Lee

In this paper, we introduce Video Region Attention Graph Networks (VRAG) that improves the state-of-the-art of video-level methods.

Retrieval Video Retrieval

Incremental-DETR: Incremental Few-Shot Object Detection via Self-Supervised Learning

no code implementations9 May 2022 Na Dong, Yongqiang Zhang, Mingli Ding, Gim Hee Lee

Incremental few-shot object detection aims at detecting novel classes without forgetting knowledge of the base classes with only a few labeled training data from the novel classes.

Few-Shot Object Detection Knowledge Distillation +3

Style-Hallucinated Dual Consistency Learning for Domain Generalized Semantic Segmentation

2 code implementations6 Apr 2022 Yuyang Zhao, Zhun Zhong, Na Zhao, Nicu Sebe, Gim Hee Lee

Furthermore, we present a novel style hallucination module (SHM) to generate style-diversified samples that are essential to consistency learning.

Domain Generalization Hallucination +3

REGTR: End-to-end Point Cloud Correspondences with Transformers

1 code implementation CVPR 2022 Zi Jian Yew, Gim Hee Lee

Despite recent success in incorporating learning into point cloud registration, many works focus on learning feature descriptors and continue to rely on nearest-neighbor feature matching and outlier filtering through RANSAC to obtain the final set of correspondences for pose estimation.

Point Cloud Registration Pose Estimation

Weakly Supervised Learning of Keypoints for 6D Object Pose Estimation

no code implementations7 Mar 2022 Meng Tian, Gim Hee Lee

Specifically, we assign a set of arbitrarily chosen 3D keypoints to represent each unknown target 3D object and learn a network to detect their 2D projections that comply with the relative camera viewpoints.

6D Pose Estimation using RGB Keypoint Detection +2

Static-Dynamic Co-Teaching for Class-Incremental 3D Object Detection

no code implementations14 Dec 2021 Na Zhao, Gim Hee Lee

Deep learning-based approaches have shown remarkable performance in the 3D object detection task.

3D Object Detection Incremental Learning +2

Novel Class Discovery in Semantic Segmentation

1 code implementation CVPR 2022 Yuyang Zhao, Zhun Zhong, Nicu Sebe, Gim Hee Lee

We introduce a new setting of Novel Class Discovery in Semantic Segmentation (NCDSS), which aims at segmenting unlabeled images containing new classes given prior knowledge from a labeled set of disjoint classes.

Image Classification Novel Class Discovery +3

Coarse-to-fine Animal Pose and Shape Estimation

1 code implementation NeurIPS 2021 Chen Li, Gim Hee Lee

Most existing animal pose and shape estimation approaches reconstruct animal meshes with a parametric SMAL model.

Learning Iterative Robust Transformation Synchronization

1 code implementation1 Nov 2021 Zi Jian Yew, Gim Hee Lee

Transformation Synchronization is the problem of recovering absolute transformations from a given set of pairwise relative motions.

Bridging Non Co-occurrence with Unlabeled In-the-wild Data for Incremental Object Detection

1 code implementation NeurIPS 2021 Na Dong, Yongqiang Zhang, Mingli Ding, Gim Hee Lee

In view of this limitation, we consider a more practical setting of complete absence of co-occurrence of the base and novel classes for the object detection task.

Incremental Learning Object +2

Voxel-based Network for Shape Completion by Leveraging Edge Generation

1 code implementation ICCV 2021 Xiaogang Wang, Marcelo H Ang Jr, Gim Hee Lee

Deep learning technique has yielded significant improvements in point cloud completion with the aim of completing missing object shapes from partial inputs.

Point Cloud Completion

Source-Free Open Compound Domain Adaptation in Semantic Segmentation

1 code implementation7 Jun 2021 Yuyang Zhao, Zhun Zhong, Zhiming Luo, Gim Hee Lee, Nicu Sebe

Second, CPSS can reduce the influence of noisy pseudo-labels and also avoid the model overfitting to the target domain during self-supervised learning, consistently boosting the performance on the target and open domains.

Domain Generalization Self-Supervised Learning +1

FILTRA: Rethinking Steerable CNN by Filter Transform

1 code implementation25 May 2021 Bo Li, Qili Wang, Gim Hee Lee

It has been an intuitive and widely used technique to construct a steerable filter by augmenting a filter with its transformed copies in the past decades, which is named as filter transform in this paper.

DeepI2P: Image-to-Point Cloud Registration via Deep Classification

1 code implementation CVPR 2021 Jiaxin Li, Gim Hee Lee

This paper presents DeepI2P: a novel approach for cross-modality registration between an image and a point cloud.

Classification General Classification +1

Multi-View Multi-Person 3D Pose Estimation with Plane Sweep Stereo

1 code implementation CVPR 2021 Jiahao Lin, Gim Hee Lee

Existing approaches for multi-view multi-person 3D pose estimation explicitly establish cross-view correspondences to group 2D pose detections from multiple camera views and solve for the 3D pose estimation for each person.

3D Multi-Person Pose Estimation 3D Pose Estimation +2

Learning Spatial Context with Graph Neural Network for Multi-Person Pose Grouping

1 code implementation6 Apr 2021 Jiahao Lin, Gim Hee Lee

More specifically, we design a Geometry-aware Association GNN that utilizes spatial information of the keypoints and learns local affinity from the global context.

Clustering graph partitioning +2

MINE: Towards Continuous Depth MPI with NeRF for Novel View Synthesis

1 code implementation ICCV 2021 Jiaxin Li, Zijian Feng, Qi She, Henghui Ding, Changhu Wang, Gim Hee Lee

In this paper, we propose MINE to perform novel view synthesis and depth estimation via dense 3D reconstruction from a single image.

3D Reconstruction Depth Estimation +1

From Synthetic to Real: Unsupervised Domain Adaptation for Animal Pose Estimation

1 code implementation CVPR 2021 Chen Li, Gim Hee Lee

Existing works circumvent this problem with pseudo labels generated from data of other easily accessible domains such as synthetic data.

Animal Pose Estimation Pseudo Label +1

City-scale Scene Change Detection using Point Clouds

no code implementations26 Mar 2021 Zi Jian Yew, Gim Hee Lee

We propose a method for detecting structural changes in a city using images captured from vehicular mounted cameras over traversals at two different times.

Change Detection Scene Change Detection

Cascaded Refinement Network for Point Cloud Completion with Self-supervision

1 code implementation17 Oct 2020 Xiaogang Wang, Marcelo H Ang Jr, Gim Hee Lee

This is to mitigate the dependence of existing approaches on large amounts of ground truth training data that are often difficult to obtain in real-world applications.

3D Object Classification Point Cloud Completion

Weakly Supervised Generative Network for Multiple 3D Human Pose Hypotheses

1 code implementation13 Aug 2020 Chen Li, Gim Hee Lee

In this paper, we propose a weakly supervised deep generative network to address the inverse problem and circumvent the need for ground truth 2D-to-3D correspondences.

Multi-Hypotheses 3D Human Pose Estimation

Point Cloud Completion by Learning Shape Priors

1 code implementation2 Aug 2020 Xiaogang Wang, Marcelo H. Ang Jr, Gim Hee Lee

Then we learn a mapping to transfer the point features from partial points to that of the complete points by optimizing feature alignment losses.

Generative Adversarial Network Point Cloud Completion

HDNet: Human Depth Estimation for Multi-Person Camera-Space Localization

1 code implementation ECCV 2020 Jiahao Lin, Gim Hee Lee

Current works on multi-person 3D pose estimation mainly focus on the estimation of the 3D joint locations relative to the root joint and ignore the absolute locations of each pose.

3D Multi-Person Pose Estimation (absolute) 3D Multi-Person Pose Estimation (root-relative) +3

Shape Prior Deformation for Categorical 6D Object Pose and Size Estimation

2 code implementations ECCV 2020 Meng Tian, Marcelo H. Ang Jr, Gim Hee Lee

We present a novel learning approach to recover the 6D poses and sizes of unseen object instances from an RGB-D image.

Object

Relative Pose Estimation of Calibrated Cameras with Known $\mathrm{SE}(3)$ Invariants

no code implementations15 Jul 2020 Bo Li, Evgeniy Martyushev, Gim Hee Lee

In this paper, we present a complete comprehensive study of the relative pose estimation problem for a calibrated camera constrained by known $\mathrm{SE}(3)$ invariant, which involves 5 minimal problems in total.

Pose Estimation Translation

Few-shot 3D Point Cloud Semantic Segmentation

1 code implementation CVPR 2021 Na Zhao, Tat-Seng Chua, Gim Hee Lee

These fully supervised approaches heavily rely on large amounts of labeled training data that are difficult to obtain and cannot segment new classes after training.

Few-shot 3D Point Cloud Semantic Segmentation Segmentation +1

Weakly Supervised Semantic Point Cloud Segmentation:Towards 10X Fewer Labels

1 code implementation8 Apr 2020 Xun Xu, Gim Hee Lee

Point cloud analysis has received much attention recently; and segmentation is one of the most important tasks.

Point Cloud Segmentation Segmentation

RPM-Net: Robust Point Matching using Learned Features

5 code implementations CVPR 2020 Zi Jian Yew, Gim Hee Lee

The hard assignments of closest point correspondences based on spatial distances are sensitive to the initial rigid transformation and noisy/outlier points, which often cause ICP to converge to wrong local minima.

Point Cloud Registration

SESS: Self-Ensembling Semi-Supervised 3D Object Detection

1 code implementation CVPR 2020 Na Zhao, Tat-Seng Chua, Gim Hee Lee

The performance of existing point cloud-based 3D object detection methods heavily relies on large-scale high-quality 3D annotations.

3D Object Detection Object +2

Identifying through Flows for Recovering Latent Representations

2 code implementations ICLR 2020 Shen Li, Bryan Hooi, Gim Hee Lee

Yet, most deep generative models do not address the question of identifiability, and thus fail to deliver on the promise of the recovery of the true latent sources that generate the observations.

Representation Learning

Trajectory Space Factorization for Deep Video-Based 3D Human Pose Estimation

1 code implementation22 Aug 2019 Jiahao Lin, Gim Hee Lee

Although existing CNN-based temporal frameworks attempt to address the sensitivity and drift problems by concurrently processing all input frames in the sequence, the existing state-of-the-art CNN-based framework is limited to 3d pose estimation of a single frame from a sequential input.

3D Pose Estimation Monocular 3D Human Pose Estimation

PS^2-Net: A Locally and Globally Aware Network for Point-Based Semantic Segmentation

1 code implementation15 Aug 2019 Na Zhao, Tat-Seng Chua, Gim Hee Lee

In this paper, we present the PS^2-Net -- a locally and globally aware deep learning framework for semantic segmentation on 3D scene-level point clouds.

Scene Segmentation Segmentation

PointAtrousGraph: Deep Hierarchical Encoder-Decoder with Point Atrous Convolution for Unorganized 3D Points

1 code implementation23 Jul 2019 Liang Pan, Chee-Meng Chew, Gim Hee Lee

Motivated by the success of encoding multi-scale contextual information for image analysis, we propose our PointAtrousGraph (PAG) - a deep permutation-invariant hierarchical encoder-decoder for efficiently exploiting multi-scale edge features in point clouds.

Robust Point Cloud Based Reconstruction of Large-Scale Outdoor Scenes

1 code implementation CVPR 2019 Ziquan Lan, Zi Jian Yew, Gim Hee Lee

Furthermore, we show that by using a Gaussian-Uniform mixture model, our approach degenerates to the formulation of a state-of-the-art approach for robust indoor reconstruction.

3D Reconstruction

Transferable Semi-supervised 3D Object Detection from RGB-D Data

no code implementations ICCV 2019 Yew Siang Tang, Gim Hee Lee

We investigate the direction of training a 3D object detector for new object classes from only 2D bounding box labels of these new classes, while simultaneously transferring information from 3D bounding box labels of the existing classes.

3D Object Detection Object +1

2D3D-MatchNet: Learning to Match Keypoints Across 2D Image and 3D Point Cloud

no code implementations22 Apr 2019 Mengdan Feng, Sixing Hu, Marcelo Ang, Gim Hee Lee

Large-scale point cloud generated from 3D sensors is more accurate than its image-based counterpart.

Pose Estimation

Discrete Rotation Equivariance for Point Cloud Recognition

1 code implementation31 Mar 2019 Jiaxin Li, Yingcai Bi, Gim Hee Lee

In this paper, we propose a deep learning architecture that achieves discrete $\mathbf{SO}(2)$/$\mathbf{SO}(3)$ rotation equivariance for point cloud recognition.

USIP: Unsupervised Stable Interest Point Detection from 3D Point Clouds

1 code implementation ICCV 2019 Jiaxin Li, Gim Hee Lee

In this paper, we propose the USIP detector: an Unsupervised Stable Interest Point detector that can detect highly repeatable and accurately localized keypoints from 3D point clouds under arbitrary transformations without the need for any ground truth training data.

Interest Point Detection

Rolling-Shutter-Aware Differential SfM and Image Rectification

no code implementations ICCV 2017 Bingbing Zhuang, Loong-Fah Cheong, Gim Hee Lee

We demonstrate that the dense depth maps recovered from the relative pose of the RS camera can be used in a RS-aware warping for image rectification to recover high-quality Global Shutter (GS) images.

3D Reconstruction Optical Flow Estimation +1

Image-Based Geo-Localization Using Satellite Imagery

no code implementations1 Mar 2019 Sixing Hu, Gim Hee Lee

The problem of localization on a geo-referenced satellite map given a query ground view image is useful yet remains challenging due to the drastic change in viewpoint.

Baseline Desensitizing In Translation Averaging

no code implementations CVPR 2018 Bingbing Zhuang, Loong-Fah Cheong, Gim Hee Lee

Many existing translation averaging algorithms are either sensitive to disparate camera baselines and have to rely on extensive preprocessing to improve the observed Epipolar Geometry graph, or if they are robust against disparate camera baselines, require complicated optimization to minimize the highly nonlinear angular error objective.

Translation

3DFeat-Net: Weakly Supervised Local 3D Features for Point Cloud Registration

1 code implementation ECCV 2018 Zi Jian Yew, Gim Hee Lee

In this paper, we propose the 3DFeat-Net which learns both 3D feature detector and descriptor for point cloud matching using weak supervision.

Point Cloud Registration

CVM-Net: Cross-View Matching Network for Image-Based Ground-to-Aerial Geo-Localization

1 code implementation CVPR 2018 Sixing Hu, Mengdan Feng, Rang M. H. Nguyen, Gim Hee Lee

The problem of localization on a geo-referenced aerial/satellite map given a query ground view image remains challenging due to the drastic change in viewpoint that causes traditional image descriptors based matching to fail.

Benchmarking Metric Learning

Solving the Perspective-2-Point Problem for Flying-Camera Photo Composition

no code implementations CVPR 2018 Ziquan Lan, David Hsu, Gim Hee Lee

The user, instead of holding a camera in hand and manually searching for a viewpoint, will interact directly with image contents in the viewfinder through simple gestures, and the flying camera will achieve the desired viewpoint through the autonomous flying capability of the drone.

PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition

5 code implementations CVPR 2018 Mikaela Angelina Uy, Gim Hee Lee

This is largely due to the difficulty in extracting local feature descriptors from a point cloud that can subsequently be encoded into a global descriptor for the retrieval task.

3D Place Recognition Point Cloud Retrieval +2

Sparse to Dense 3D Reconstruction From Rolling Shutter Images

no code implementations CVPR 2016 Olivier Saurer, Marc Pollefeys, Gim Hee Lee

It is well known that the rolling shutter effect in images captured with a moving rolling shutter camera causes inaccuracies to 3D reconstructions.

3D Reconstruction

Relative Pose Estimation for a Multi-Camera System with Known Vertical Direction

no code implementations CVPR 2014 Gim Hee Lee, Marc Pollefeys, Friedrich Fraundorfer

In this paper, we present our minimal 4-point and linear 8-point algorithms to estimate the relative pose of a multi-camera system with known vertical directions, i. e. known absolute roll and pitch angles.

Pose Estimation

Motion Estimation for Self-Driving Cars with a Generalized Camera

no code implementations CVPR 2013 Gim Hee Lee, Friedrich Faundorfer, Marc Pollefeys

By modeling the multicamera system as a generalized camera and applying the non-holonomic motion constraint of a car, we show that this leads to a novel 2-point minimal solution for the generalized essential matrix where the full relative motion including metric scale can be obtained.

Motion Estimation Self-Driving Cars

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