Filters








2,470 Hits in 5.2 sec

User-Specific Hand Modeling from Monocular Depth Sequences

Jonathan Taylor, Richard Stebbing, Varun Ramakrishna, Cem Keskin, Jamie Shotton, Shahram Izadi, Aaron Hertzmann, Andrew Fitzgibbon
2014 2014 IEEE Conference on Computer Vision and Pattern Recognition  
This paper presents a method for acquiring dense nonrigid shape and deformation from a single monocular depth sensor.  ...  We combine and extend existing work on model-based tracking, subdivision surface fitting, and mesh deformation to acquire detailed hand models from as few as 15 frames of depth data.  ...  Our new optimization framework can reliably infer a user-specific hand mesh from a coarse rigged template using as few as 15 noisy frames from a commodity depth sensor as input.  ... 
doi:10.1109/cvpr.2014.88 dblp:conf/cvpr/TaylorSRKSIHF14 fatcat:2mqmgjy6w5azti7qdwkfmvigku

SLAM in the Field: An Evaluation of Monocular Mapping and Localization on Challenging Dynamic Agricultural Environment [article]

Fangwen Shu, Paul Lesur, Yaxu Xie, Alain Pagani, Didier Stricker
2020 arXiv   pre-print
To the best of our knowledge, this paper presents the first evaluation results for monocular SLAM, and our work further explores unsupervised depth estimation on this specific application scenario by simulating  ...  Moreover, we highlight that our experiments provide meaningful insight to improve monocular SLAM systems under agricultural settings.  ...  However, pretrained model guarantees the stability and robustness of RGB-D based tracking, while tracking failure continues to happen on all the sequence using the model specifically from monocular training  ... 
arXiv:2011.01122v2 fatcat:myxe7aoaszepxij37q6t522edq

GANerated Hands for Real-time 3D Hand Tracking from Monocular RGB [article]

Franziska Mueller, Florian Bernard, Oleksandr Sotnychenko, Dushyant Mehta, Srinath Sridhar, Dan Casas, Christian Theobalt
2017 arXiv   pre-print
We address the highly challenging problem of real-time 3D hand tracking based on a monocular RGB-only sequence.  ...  Our tracking method combines a convolutional neural network with a kinematic 3D hand model, such that it generalizes well to unseen data, is robust to occlusions and varying camera viewpoints, and leads  ...  The idea of using user-specific positions is to avoid mispredictions caused by bone length inconsistencies in the hand model by normalizing the predictions to comply with the model.  ... 
arXiv:1712.01057v1 fatcat:7jxgkuoogfbb5itjoq2wbpftpa

Garment Replacement in Monocular Video Sequences

Lorenz Rogge, Felix Klose, Michael Stengel, Martin Eisemann, Marcus Magnor
2014 ACM Transactions on Graphics  
We reconstruct the actor's body shape and motion from the input video using a parameterized body model.  ...  We present a semi-automatic approach to exchange the clothes of an actor for arbitrary virtual garments in conventional monocular video footage as a post-process.  ...  ACKNOWLEDGMENTS The research leading to these results has received funding from the European Unions Seventh Framework Programme FP7/2007-2013 under grant agreement no. 256941, Reality CG.  ... 
doi:10.1145/2634212 fatcat:crudba25orabpi4ihtz4fnf4de

GANerated Hands for Real-Time 3D Hand Tracking from Monocular RGB

Franziska Mueller, Florian Bernard, Oleksandr Sotnychenko, Dushyant Mehta, Srinath Sridhar, Dan Casas, Christian Theobalt
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Figure 1 : We present an approach for real-time 3D hand tracking from monocular RGB-only input.  ...  Abstract We address the highly challenging problem of real-time 3D hand tracking based on a monocular RGB-only sequence.  ...  The idea of using user-specific positions is to avoid mispredictions caused by bone length inconsistencies in the hand model by normalizing the predictions to comply with the model.  ... 
doi:10.1109/cvpr.2018.00013 dblp:conf/cvpr/MuellerBSM0CT18 fatcat:pw73umrjgjhdzpttehh3uljeu4

Self-supervised learning for autonomous vehicles perception: A conciliation between analytical and learning methods [article]

Florent Chiaroni, Mohamed-Cherif Rahal, Nicolas Hueber, Frederic Dufaux
2020 arXiv   pre-print
alternative to supervised learning for a variety of different tasks, including long-range traversable area segmentation, moving obstacle instance segmentation, long-term moving obstacle tracking, or depth  ...  [54] propose to predict, from a temporal sequence of frames, the depth map with a learning model, and the successive camera pose transformations with another learning model.  ...  approach presented in [16] predicts a depth map from a monocular camera without relying on annotated depth maps.  ... 
arXiv:1910.01636v2 fatcat:6z55krsbjve65pvzfnaidfo6ca

MonoMR: Synthesizing Pseudo-2.5D Mixed Reality Content from Monocular Videos

Dong-Hyun Hwang, Hideki Koike
2021 Applied Sciences  
MonoMR is a system that synthesizes pseudo-2.5D content from monocular videos for mixed reality (MR) head-mounted displays (HMDs).  ...  Unlike conventional systems that require multiple cameras, the MonoMR system can be used by casual end-users to generate MR content from a single camera only.  ...  Hence, it allows the user to select and enjoy scenes from a specific viewpoint.  ... 
doi:10.3390/app11177946 fatcat:hdvq5gay65hmha2y4745kavx3a

Lightweight Markerless Monocular Face Capture with 3D Spatial Priors [article]

Shridhar Ravikumar
2019 arXiv   pre-print
shapes into an integrated objective function. 2D monocular inputs inherently lack information along the depth axis and can lead to physically implausible solutions.  ...  We present a simple lightweight markerless facial performance capture framework using just a monocular video input that combines Active Appearance Models for feature tracking and prior constraints on 3D  ...  On the other hand, parametric model based approaches optimize for the values of the parameters through the sequence in order to best capture the movement of the face.  ... 
arXiv:1901.05355v1 fatcat:2cpchmuq5vhkxf55qikegwqrrm

Learning Marginalization through Regression for Hand Orientation Inference

Muhammad Asad, Greg Slabaugh
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
The proposed model is composed of two layers, where the first layer consists of a marginalization weights regressor while the second layer contains expert regressors trained on subsets of our hand orientation  ...  Our main contribution comes from the regression based marginalization of these posterior probabilities.  ...  This depicts an application scenario where a one-time model calibration will require the user to provide a user-specific hand orientation dataset.  ... 
doi:10.1109/cvprw.2016.154 dblp:conf/cvpr/AsadS16 fatcat:nttxoronhjcidhek7zuwtbxmqy

UnrealROX: an extremely photorealistic virtual reality environment for robotics simulations and synthetic data generation

Pablo Martinez-Gonzalez, Sergiu Oprea, Alberto Garcia-Garcia, Alvaro Jover-Alvarez, Sergio Orts-Escolano, Jose Garcia-Rodriguez
2019 Virtual Reality  
However, most of the current datasets and environments lack realism, interactions, and details from the real world.  ...  Photorealistic scenes and robots are rendered by Unreal Engine into a virtual reality headset which captures gaze so that a human operator can move the robot and use controllers for the robotic hands;  ...  Experiments were made possible by a generous hardware donation from NVIDIA. We would also like to thank Zuria Bauer for her collaboration in the depth estimation experiments.  ... 
doi:10.1007/s10055-019-00399-5 fatcat:jkrft36t5zcqnke4r7qst7xpha

UnrealROX: An eXtremely Photorealistic Virtual Reality Environment for Robotics Simulations and Synthetic Data Generation [article]

Pablo Martinez-Gonzalez, Sergiu Oprea, Alberto Garcia-Garcia, Alvaro Jover-Alvarez, Sergio Orts-Escolano, Jose Garcia-Rodriguez
2019 arXiv   pre-print
However, most of the current datasets and environments lack realism, interactions, and details from the real world.  ...  Photorealistic scenes and robots are rendered by Unreal Engine into a virtual reality headset which captures gaze so that a human operator can move the robot and use controllers for the robotic hands;  ...  Experiments were made possible by a generous hardware donation from NVIDIA. We would also like to thank Zuria Bauer for her collaboration in the depth estimation experiments.  ... 
arXiv:1810.06936v2 fatcat:3e4vionszbhfjetr2zzrbd4uxu

Weakly Supervised Training of Monocular 3D Object Detectors Using Wide Baseline Multi-view Traffic Camera Data [article]

Matthew Howe, Ian Reid, Jamie Mackenzie
2021 arXiv   pre-print
To fine-tune the monocular 3D object detector, our method utilises multiple 2D detections from overlapping, wide-baseline views and a loss that encodes the subjacent geometric consistency.  ...  Accurate 7DoF prediction of vehicles at an intersection is an important task for assessing potential conflicts between road users.  ...  Specifically, our method allows a road safety researcher to fine-tune a model for a specific camera set up with 10 minutes of additional video data from cameras with overlapping fields of view and no hand  ... 
arXiv:2110.10966v1 fatcat:angwnguodbcfblpblycu6gorzq

RGB2Hands: Real-Time Tracking of 3D Hand Interactions from Monocular RGB Video [article]

Jiayi Wang, Franziska Mueller, Florian Bernard, Suzanne Sorli, Oleksandr Sotnychenko, Neng Qian, Miguel A. Otaduy, Dan Casas, Christian Theobalt
2021 arXiv   pre-print
a 3D hand model, and 2D keypoint positions, together with newly proposed intra-hand relative depth and inter-hand distance maps.  ...  These predictions are subsequently used in a generative model fitting framework in order to estimate pose and shape parameters of a 3D hand model for both hands.  ...  very close two-hand interactions from monocular RGB images.  ... 
arXiv:2106.11725v1 fatcat:esmiakdncjco3cb6uhdzaqosoq

Direct, Dense, and Deformable: Template-Based Non-rigid 3D Reconstruction from RGB Video

Rui Yu, Chris Russell, Neill D. F. Campbell, Lourdes Agapito
2015 2015 IEEE International Conference on Computer Vision (ICCV)  
channel available in RGB-D cameras, or deal with specific shapes such as faces or planar surfaces.  ...  We first compute a dense 3D template of the shape of the object, using a short rigid sequence, and subsequently perform online reconstruction of the non-rigid mesh as it evolves over time.  ...  This work has been partly supported by the Second-Hands project, funded from the European Unions Horizon 2020 Research and Innovation programme under grant agreement No 643950.  ... 
doi:10.1109/iccv.2015.111 dblp:conf/iccv/YuRCA15 fatcat:gjutjeyyt5gkhhe3mmcwfuyhk4

3D Reconstruction of Human Skeleton from Single Images or Monocular Video Sequences [chapter]

Fabio Remondino, Andreas Roditakis
2003 Lecture Notes in Computer Science  
video sequence.  ...  In this paper, we first review the approaches to recover 3D shape and related movements of a human and then we present an easy and reliable approach to recover a 3D model using just one image or monocular  ...  Conclusion In this work we presented the problem of recovering 3D models of humans from single images and monocular video sequences.  ... 
doi:10.1007/978-3-540-45243-0_14 fatcat:ic5rpf3tnzaibnwudcdef3wmdi
« Previous Showing results 1 — 15 out of 2,470 results