Search Results for author: Tae-Kyun Kim

Found 74 papers, 14 papers with code

Unsupervised Learning of Optical Flow with Deep Feature Similarity

no code implementations ECCV 2020 Woobin Im, Tae-Kyun Kim, Sung-Eui Yoon

Deep unsupervised learning for optical flow has been proposed, where the loss measures image similarity with the warping function parameterized by estimated flow.

Optical Flow Estimation

A Unified Architecture of Semantic Segmentation and Hierarchical Generative Adversarial Networks for Expression Manipulation

no code implementations8 Dec 2021 Rumeysa Bodur, Binod Bhattarai, Tae-Kyun Kim

Recently, hierarchical networks that consist of both a global network dealing with the whole image and multiple local networks focusing on local parts are showing success.

Facial Expression Translation Image Manipulation +1

EvDistill: Asynchronous Events to End-task Learning via Bidirectional Reconstruction-guided Cross-modal Knowledge Distillation

1 code implementation CVPR 2021 Lin Wang, Yujeong Chae, Sung-Hoon Yoon, Tae-Kyun Kim, Kuk-Jin Yoon

To enable KD across the unpaired modalities, we first propose a bidirectional modality reconstruction (BMR) module to bridge both modalities and simultaneously exploit them to distill knowledge via the crafted pairs, causing no extra computation in the inference.

Knowledge Distillation Object Recognition +1

SeCGAN: Parallel Conditional Generative Adversarial Networks for Face Editing via Semantic Consistency

no code implementations17 Nov 2021 Jiaze Sun, Binod Bhattarai, Zhixiang Chen, Tae-Kyun Kim

Whilst both branches are required during training, the RGB branch is our primary network and the semantic branch is not needed for inference.

Visual Transformer for Task-aware Active Learning

1 code implementation7 Jun 2021 Razvan Caramalau, Binod Bhattarai, Tae-Kyun Kim

In this paper, we present a novel pipeline for pool-based Active Learning.

Active Learning

SHAQ: Incorporating Shapley Value Theory into Multi-Agent Q-Learning

1 code implementation31 May 2021 Jianhong Wang, Yuan Zhang, Yunjie Gu, Tae-Kyun Kim

This paper studies a theoretical framework for value factorisation with interpretability via Shapley value theory.

Fairness Q-Learning +2

Label Geometry Aware Discriminator for Conditional Generative Networks

no code implementations12 May 2021 Suman Sapkota, Bidur Khanal, Binod Bhattarai, Bishesh Khanal, Tae-Kyun Kim

Multi-domain image-to-image translation with conditional Generative Adversarial Networks (GANs) can generate highly photo realistic images with desired target classes, yet these synthetic images have not always been helpful to improve downstream supervised tasks such as image classification.

Data Augmentation Image Classification +1

Learning Feature Aggregation for Deep 3D Morphable Models

1 code implementation CVPR 2021 Zhixiang Chen, Tae-Kyun Kim

3D morphable models are widely used for the shape representation of an object class in computer vision and graphics applications.

Geometry-based Distance Decomposition for Monocular 3D Object Detection

1 code implementation ICCV 2021 Xuepeng Shi, Qi Ye, Xiaozhi Chen, Chuangrong Chen, Zhixiang Chen, Tae-Kyun Kim

The experimental results show that our method achieves the state-of-the-art performance on the monocular 3D Object Detection and Birds Eye View tasks of the KITTI dataset, and can generalize to images with different camera intrinsics.

Autonomous Driving Monocular 3D Object Detection +1

Adversarial Imitation Learning with Trajectorial Augmentation and Correction

no code implementations25 Mar 2021 Dafni Antotsiou, Carlo Ciliberto, Tae-Kyun Kim

Deep Imitation Learning requires a large number of expert demonstrations, which are not always easy to obtain, especially for complex tasks.

Data Augmentation Imitation Learning

Active Learning for Bayesian 3D Hand Pose Estimation

1 code implementation1 Oct 2020 Razvan Caramalau, Binod Bhattarai, Tae-Kyun Kim

We propose a Bayesian approximation to a deep learning architecture for 3D hand pose estimation.

3D Hand Pose Estimation Active Learning

3D Dense Geometry-Guided Facial Expression Synthesis by Adversarial Learning

no code implementations30 Sep 2020 Rumeysa Bodur, Binod Bhattarai, Tae-Kyun Kim

We utilise this dataset to minimise the novel depth consistency loss via adversarial learning (note we do not have ground truth depth maps for generated face images) and the depth categorical loss of synthetic data on the discriminator.

3D Reconstruction

Physics-Based Dexterous Manipulations with Estimated Hand Poses and Residual Reinforcement Learning

no code implementations7 Aug 2020 Guillermo Garcia-Hernando, Edward Johns, Tae-Kyun Kim

Dexterous manipulation of objects in virtual environments with our bare hands, by using only a depth sensor and a state-of-the-art 3D hand pose estimator (HPE), is challenging.

3D Hand Pose Estimation Imitation Learning +1

Sequential Graph Convolutional Network for Active Learning

1 code implementation CVPR 2021 Razvan Caramalau, Binod Bhattarai, Tae-Kyun Kim

We flip the label of newly queried nodes from unlabelled to labelled, re-train the learner to optimise the downstream task and the graph to minimise its modified objective.

Active Learning Hand Pose Estimation +1

MatchGAN: A Self-Supervised Semi-Supervised Conditional Generative Adversarial Network

no code implementations11 Jun 2020 Jiaze Sun, Binod Bhattarai, Tae-Kyun Kim

We perform augmentation by randomly sampling sensible labels from the label space of the few labelled examples available and assigning them as target labels to the abundant unlabelled examples from the same distribution as that of the labelled ones.

Self-Supervised Learning

Modelling Hierarchical Structure between Dialogue Policy and Natural Language Generator with Option Framework for Task-oriented Dialogue System

1 code implementation ICLR 2021 Jianhong Wang, Yuan Zhang, Tae-Kyun Kim, Yunjie Gu

We test HDNO on MultiWoz 2. 0 and MultiWoz 2. 1, the datasets on multi-domain dialogues, in comparison with word-level E2E model trained with RL, LaRL and HDSA, showing improvements on the performance evaluated by automatic evaluation metrics and human evaluation.

Hierarchical Reinforcement Learning reinforcement-learning +1

Introducing Pose Consistency and Warp-Alignment for Self-Supervised 6D Object Pose Estimation in Color Images

no code implementations27 Mar 2020 Juil Sock, Guillermo Garcia-Hernando, Anil Armagan, Tae-Kyun Kim

Most successful approaches to estimate the 6D pose of an object typically train a neural network by supervising the learning with annotated poses in real world images.

6D Pose Estimation using RGB Domain Adaptation +1

Additive Angular Margin for Few Shot Learning to Classify Clinical Endoscopy Images

no code implementations23 Mar 2020 Sharib Ali, Binod Bhattarai, Tae-Kyun Kim, Jens Rittscher

In this work, we propose to use a few-shot learning approach that requires less training data and can be used to predict label classes of test samples from an unseen dataset.

Few-Shot Learning

EventSR: From Asynchronous Events to Image Reconstruction, Restoration, and Super-Resolution via End-to-End Adversarial Learning

1 code implementation CVPR 2020 Lin Wang, Tae-Kyun Kim, Kuk-Jin Yoon

While each phase is mainly for one of the three tasks, the networks in earlier phases are fine-tuned by respective loss functions in an end-to-end manner.

Image Reconstruction Super-Resolution

Inducing Optimal Attribute Representations for Conditional GANs

no code implementations ECCV 2020 Binod Bhattarai, Tae-Kyun Kim

Existing conditional GANs commonly encode target domain label information as hard-coded categorical vectors in the form of 0s and 1s.

Multi-Task Learning SSIM

A Review on Object Pose Recovery: from 3D Bounding Box Detectors to Full 6D Pose Estimators

no code implementations28 Jan 2020 Caner Sahin, Guillermo Garcia-Hernando, Juil Sock, Tae-Kyun Kim

In this paper, we present the first comprehensive and most recent review of the methods on object pose recovery, from 3D bounding box detectors to full 6D pose estimators.

6D Pose Estimation using RGB Autonomous Driving +1

Accurate 6D Object Pose Estimation by Pose Conditioned Mesh Reconstruction

no code implementations23 Oct 2019 Pedro Castro, Anil Armagan, Tae-Kyun Kim

Current 6D object pose methods consist of deep CNN models fully optimized for a single object but with its architecture standardized among objects with different shapes.

6D Pose Estimation using RGB

Active 6D Multi-Object Pose Estimation in Cluttered Scenarios with Deep Reinforcement Learning

no code implementations19 Oct 2019 Juil Sock, Guillermo Garcia-Hernando, Tae-Kyun Kim

In this work, we explore how a strategic selection of camera movements can facilitate the task of 6D multi-object pose estimation in cluttered scenarios while respecting real-world constraints important in robotics and augmented reality applications, such as time and distance traveled.

Pose Estimation reinforcement-learning

Sampling Strategies for GAN Synthetic Data

no code implementations10 Sep 2019 Binod Bhattarai, Seungryul Baek, Rumeysa Bodur, Tae-Kyun Kim

Unlike previous studies of randomly augmenting the synthetic data with real data, we present our simple, effective and easy to implement synthetic data sampling methods to train deep CNNs more efficiently and accurately.

Real-time Background-aware 3D Textureless Object Pose Estimation

no code implementations22 Jul 2019 Mang Shao, Danhang Tang, Tae-Kyun Kim

In this work, we present a modified fuzzy decision forest for real-time 3D object pose estimation based on typical template representation.

Pose Estimation

Shapley Q-value: A Local Reward Approach to Solve Global Reward Games

2 code implementations11 Jul 2019 Jianhong Wang, Yuan Zhang, Tae-Kyun Kim, Yunjie Gu

To deal with this problem, we i) introduce a cooperative-game theoretical framework called extended convex game (ECG) that is a superset of global reward game, and ii) propose a local reward approach called Shapley Q-value.

Multi-agent Reinforcement Learning Policy Gradient Methods

Pushing the Envelope for RGB-based Dense 3D Hand Pose Estimation via Neural Rendering

no code implementations CVPR 2019 Seungryul Baek, Kwang In Kim, Tae-Kyun Kim

Once the model is successfully fitted to input RGB images, its meshes i. e. shapes and articulations, are realistic, and we augment view-points on top of estimated dense hand poses.

3D Hand Pose Estimation 3D Pose Estimation +2

Instance- and Category-level 6D Object Pose Estimation

no code implementations11 Mar 2019 Caner Sahin, Guillermo Garcia-Hernando, Juil Sock, Tae-Kyun Kim

6D object pose estimation is an important task that determines the 3D position and 3D rotation of an object in camera-centred coordinates.

6D Pose Estimation using RGB Scene Understanding

RGB-based 3D Hand Pose Estimation via Privileged Learning with Depth Images

no code implementations18 Nov 2018 Shanxin Yuan, Bjorn Stenger, Tae-Kyun Kim

We explore different ways of using this privileged information: (1) using depth data to initially train a depth-based network, (2) using the features from the depth-based network of the paired depth images to constrain mid-level RGB network weights, and (3) using the foreground mask, obtained from the depth data, to suppress the responses from the background area.

3D Hand Pose Estimation

Category-level 6D Object Pose Recovery in Depth Images

no code implementations1 Aug 2018 Caner Sahin, Tae-Kyun Kim

Intra-class variations, distribution shifts among source and target domains are the major challenges of category-level tasks.

6D Pose Estimation using RGB Graph Matching

Multi-Task Deep Networks for Depth-Based 6D Object Pose and Joint Registration in Crowd Scenarios

no code implementations11 Jun 2018 Juil Sock, Kwang In Kim, Caner Sahin, Tae-Kyun Kim

Our architecture jointly learns multiple sub-tasks: 2D detection, depth, and 3D pose estimation of individual objects; and joint registration of multiple objects.

3D Pose Estimation Multi-Task Learning

Augmented Skeleton Space Transfer for Depth-based Hand Pose Estimation

no code implementations CVPR 2018 Seungryul Baek, Kwang In Kim, Tae-Kyun Kim

By training the HPG and HPE in a single unified optimization framework enforcing that 1) the HPE agrees with the paired depth and skeleton entries; and 2) the HPG-HPE combination satisfies the cyclic consistency (both the input and the output of HPG-HPE are skeletons) observed via the newly generated unpaired skeletons, our algorithm constructs a HPE which is robust to variations that go beyond the coverage of the existing database.

Hand Pose Estimation

Semi-supervised Adversarial Learning to Generate Photorealistic Face Images of New Identities from 3D Morphable Model

1 code implementation ECCV 2018 Baris Gecer, Binod Bhattarai, Josef Kittler, Tae-Kyun Kim

We propose a novel end-to-end semi-supervised adversarial framework to generate photorealistic face images of new identities with wide ranges of expressions, poses, and illuminations conditioned by a 3D morphable model.

Domain Adaptation Face Generation +3

Occlusion-aware Hand Pose Estimation Using Hierarchical Mixture Density Network

no code implementations ECCV 2018 Qi Ye, Tae-Kyun Kim

The proposed method leverages the state-of-the-art hand pose estimators based on Convolutional Neural Networks to facilitate feature learning, while it models the multiple modes in a two-level hierarchy to reconcile single-valued and multi-valued mapping in its output.

Hand Pose Estimation

Learning Deep Convolutional Embeddings for Face Representation Using Joint Sample- and Set-based Supervision

no code implementations1 Aug 2017 Baris Gecer, Vassileios Balntas, Tae-Kyun Kim

In this work, we investigate several methods and strategies to learn deep embeddings for face recognition, using joint sample- and set-based optimization.

Face Recognition

The 2017 Hands in the Million Challenge on 3D Hand Pose Estimation

no code implementations7 Jul 2017 Shanxin Yuan, Qi Ye, Guillermo Garcia-Hernando, Tae-Kyun Kim

We present the 2017 Hands in the Million Challenge, a public competition designed for the evaluation of the task of 3D hand pose estimation.

3D Hand Pose Estimation

Recovering 6D Object Pose: A Review and Multi-modal Analysis

no code implementations10 Jun 2017 Caner Sahin, Tae-Kyun Kim

A large number of studies analyse object detection and pose estimation at visual level in 2D, discussing the effects of challenges such as occlusion, clutter, texture, etc., on the performances of the methods, which work in the context of RGB modality.

6D Pose Estimation 6D Pose Estimation using RGB +2

Deep Convolutional Decision Jungle for Image Classification

no code implementations6 Jun 2017 Seungryul Baek, Kwang In Kim, Tae-Kyun Kim

Each response map-or node-in both the convolutional and fully-connected layers selectively respond to class labels s. t.

Classification Face Verification +2

First-Person Hand Action Benchmark with RGB-D Videos and 3D Hand Pose Annotations

1 code implementation CVPR 2018 Guillermo Garcia-Hernando, Shanxin Yuan, Seungryul Baek, Tae-Kyun Kim

Our dataset and experiments can be of interest to communities of 3D hand pose estimation, 6D object pose, and robotics as well as action recognition.

3D Hand Pose Estimation Action Recognition +2

A Learning-based Variable Size Part Extraction Architecture for 6D Object Pose Recovery in Depth

no code implementations9 Jan 2017 Caner Sahin, Rigas Kouskouridas, Tae-Kyun Kim

The iterative refinement is accomplished based on finer (smaller) parts that are represented with more discriminative control point descriptors by using our Iterative Hough Forest.

Real-time Online Action Detection Forests using Spatio-temporal Contexts

no code implementations28 Oct 2016 Seungryul Baek, Kwang In Kim, Tae-Kyun Kim

Online action detection (OAD) is challenging since 1) robust yet computationally expensive features cannot be straightforwardly used due to the real-time processing requirements and 2) the localization and classification of actions have to be performed even before they are fully observed.

Online Action Detection

Kinematic-Layout-aware Random Forests for Depth-based Action Recognition

no code implementations23 Jul 2016 Seungryul Baek, Zhiyuan Shi, Masato Kawade, Tae-Kyun Kim

In this paper, we tackle the problem of 24 hours-monitoring patient actions in a ward such as "stretching an arm out of the bed", "falling out of the bed", where temporal movements are subtle or significant.

Action Recognition

Transition Forests: Learning Discriminative Temporal Transitions for Action Recognition and Detection

no code implementations CVPR 2017 Guillermo Garcia-Hernando, Tae-Kyun Kim

A human action can be seen as transitions between one's body poses over time, where the transition depicts a temporal relation between two poses.

Action Detection Action Recognition +1

Siamese Regression Networks with Efficient mid-level Feature Extraction for 3D Object Pose Estimation

no code implementations8 Jul 2016 Andreas Doumanoglou, Vassileios Balntas, Rigas Kouskouridas, Tae-Kyun Kim

Furthermore, we argue that our pose-guided feature learning using our Siamese Regression Network generates more discriminative features that outperform the state of the art.

3D Pose Estimation

Latent Bi-constraint SVM for Video-based Object Recognition

no code implementations31 May 2016 Yang Liu, Minh Hoai, Mang Shao, Tae-Kyun Kim

LBSVM is based on Structured-Output SVM, but extends it to handle noisy video data and ensure consistency of the output decision throughout time.

Object Recognition

Effective Backscatter Approximation for Photometry in Murky Water

no code implementations29 Apr 2016 Chourmouzios Tsiotsios, Maria E. Angelopoulou, Andrew J. Davison, Tae-Kyun Kim

Backscatter corresponds to a complex term with several unknown variables, and makes the problem of normal estimation hard.

Image Restoration

Spatial Attention Deep Net with Partial PSO for Hierarchical Hybrid Hand Pose Estimation

no code implementations12 Apr 2016 Qi Ye, Shanxin Yuan, Tae-Kyun Kim

In this paper, a hybrid hand pose estimation method is proposed by applying the kinematic hierarchy strategy to the input space (as well as the output space) of the discriminative method by a spatial attention mechanism and to the optimization of the generative method by hierarchical Particle Swarm Optimization (PSO).

Hand Pose Estimation

Iterative Hough Forest with Histogram of Control Points for 6 DoF Object Registration from Depth Images

no code implementations8 Mar 2016 Caner Sahin, Rigas Kouskouridas, Tae-Kyun Kim

State-of-the-art techniques proposed for 6D object pose recovery depend on occlusion-free point clouds to accurately register objects in 3D space.

Pose Estimation

Latent-Class Hough Forests for 6 DoF Object Pose Estimation

no code implementations3 Feb 2016 Rigas Kouskouridas, Alykhan Tejani, Andreas Doumanoglou, Danhang Tang, Tae-Kyun Kim

In this paper we present Latent-Class Hough Forests, a method for object detection and 6 DoF pose estimation in heavily cluttered and occluded scenarios.

object-detection Object Detection +2

Recovering 6D Object Pose and Predicting Next-Best-View in the Crowd

no code implementations CVPR 2016 Andreas Doumanoglou, Rigas Kouskouridas, Sotiris Malassiotis, Tae-Kyun Kim

In this work, we present a complete framework for both single shot-based 6D object pose estimation and next-best-view prediction based on Hough Forests, the state of the art object pose estimator that performs classification and regression jointly.

6D Pose Estimation 6D Pose Estimation using RGB +2

Conditional Convolutional Neural Network for Modality-Aware Face Recognition

no code implementations ICCV 2015 Chao Xiong, Xiaowei Zhao, Danhang Tang, Karlekar Jayashree, Shuicheng Yan, Tae-Kyun Kim

Faces in the wild are usually captured with various poses, illuminations and occlusions, and thus inherently multimodally distributed in many tasks.

Face Identification Face Recognition +1

Opening the Black Box: Hierarchical Sampling Optimization for Estimating Human Hand Pose

no code implementations ICCV 2015 Danhang Tang, Jonathan Taylor, Pushmeet Kohli, Cem Keskin, Tae-Kyun Kim, Jamie Shotton

In this paper, we show that we can significantly improving upon black box optimization by exploiting high-level knowledge of the structure of the parameters and using a local surrogate energy function.

Hand Pose Estimation Image Generation

Enhanced Random Forest with Image/Patch-Level Learning for Image Understanding

no code implementations14 Oct 2014 Wai Lam Hoo, Tae-Kyun Kim, Yuru Pei, Chee Seng Chan

Image understanding is an important research domain in the computer vision due to its wide real-world applications.

Multiple Object Tracking: A Literature Review

no code implementations26 Sep 2014 Wenhan Luo, Junliang Xing, Anton Milan, Xiaoqin Zhang, Wei Liu, Tae-Kyun Kim

We inspect the recent advances in various aspects and propose some interesting directions for future research.

Multiple Object Tracking

Bi-label Propagation for Generic Multiple Object Tracking

no code implementations CVPR 2014 Wenhan Luo, Tae-Kyun Kim, Bjorn Stenger, Xiaowei Zhao, Roberto Cipolla

In this paper, we propose a label propagation framework to handle the multiple object tracking (MOT) problem for a generic object type (cf.

Multiple Object Tracking

Unified Face Analysis by Iterative Multi-Output Random Forests

no code implementations CVPR 2014 Xiaowei Zhao, Tae-Kyun Kim, Wenhan Luo

In this paper, we present a unified method for joint face image analysis, i. e., simultaneously estimating head pose, facial expression and landmark positions in real-world face images.

Backscatter Compensated Photometric Stereo with 3 Sources

no code implementations CVPR 2014 Chourmouzios Tsiotsios, Maria E. Angelopoulou, Tae-Kyun Kim, Andrew J. Davison

We compare our method with previous approaches through extensive experimental results, where a variety of objects are imaged in a big water tank whose turbidity is systematically increased, and show reconstruction quality which degrades little relative to clean water results even with a very significant scattering level.

Latent Regression Forest: Structured Estimation of 3D Articulated Hand Posture

no code implementations CVPR 2014 Danhang Tang, Hyung Jin Chang, Alykhan Tejani, Tae-Kyun Kim

In contrast to prior forest-based methods, which take dense pixels as input, classify them independently and then estimate joint positions afterwards; our method can be considered as a structured coarse-to-fine search, starting from the centre of mass of a point cloud until locating all the skeletal joints.

3D Hand Pose Estimation

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