Filters








15,719 Hits in 3.5 sec

Transferring Dense Pose to Proximal Animal Classes [article]

Artsiom Sanakoyeu, Vasil Khalidov, Maureen S. McCarthy, Andrea Vedaldi, Natalia Neverova
2020 arXiv   pre-print
We show that, at least for proximal animal classes such as chimpanzees, it is possible to transfer the knowledge existing in dense pose recognition for humans, as well as in more general object detectors  ...  and segmenters, to the problem of dense pose recognition in other classes.  ...  Encouragingly, we have demonstrated that existing detection, segmentation, and dense pose labelling models can transfer very well to a proximal animal class such as chimpanzee despite significant inter-class  ... 
arXiv:2003.00080v1 fatcat:lnwjx2m73ven5fvlavpvnpkf5i

Discovering Relationships between Object Categories via Universal Canonical Maps [article]

Natalia Neverova, Artsiom Sanakoyeu, Patrick Labatut, David Novotny, Andrea Vedaldi
2021 arXiv   pre-print
Recent work has shown that it is possible to learn a unified dense pose predictor for several categories of related objects.  ...  Moreover, the new model is also better at the task of dense pose prediction than prior work.  ...  In pose to proximal animal classes.  ... 
arXiv:2106.09758v1 fatcat:5a32sx4uirfbdpzhu2t6rqeqla

Pose Recognition in the Wild: Animal pose estimation using Agglomerative Clustering and Contrastive Learning [article]

Samayan Bhattacharya, Sk Shahnawaz
2021 arXiv   pre-print
In addition, due to the wide variety of body shapes in the animal kingdom, the transfer of knowledge across domains is ineffective.  ...  Animal pose estimation has recently come into the limelight due to its application in biology, zoology, and aquaculture. Deep learning methods have effectively been applied to human pose estimation.  ...  Transferring knowledge has been proposed to adapt the annotations existing for human images to animal images [7] .  ... 
arXiv:2111.08259v1 fatcat:6utxtcibwrfaxg2w3wjceg25hm

A framework for image-based asset generation and animation

Johannes Furch, Anna Hilsmann, Peter Eisert
2015 2015 IEEE International Conference on Image Processing (ICIP)  
Real-time photo-realistic animation is enabled by the use of captured images and shifting computational complexity to an a-priori training phase.  ...  This can only be achieved in well funded film productions, while in low-cost applications, animated objects usually lack visual quality.  ...  To simplify the data transfer to the graphics hardware, we limit the number of influencing bones for each vertex to four. Proximity Graphs.  ... 
doi:10.1109/icip.2015.7351141 dblp:conf/icip/FurchHE15 fatcat:cs2rt7yuqzcdxabizotfvy3sby

Making DensePose fast and light [article]

Ruslan Rakhimov, Emil Bogomolov, Alexandr Notchenko, Fung Mao, Alexey Artemov, Denis Zorin, Evgeny Burnaev
2020 arXiv   pre-print
To enable Dense Pose inference on the end device with current models, one needs to support an expensive server-side infrastructure and have a stable internet connection.  ...  Existing neural network models capable of solving this task are heavily parameterized and a long way from being transferred to an embedded or mobile device.  ...  [15] shows the ability to transfer the dense pose recognition from humans to proximal animal classes such as chimpanzees without a time-consuming collection of a new dataset with new classes.  ... 
arXiv:2006.15190v3 fatcat:epwsa6psobbyxmnh24b4blly4q

Structure and function of the pituitary gland in gonadal tumor-bearing and normal cyprinid fish

J F Leatherland, R A Sonstegard
1977 Cancer Research  
Tumor-bearing carp also exhibited gonadotroph hy perplasia similar to that in tumorous hybrids.  ...  The F, generation hybrids of carp and goldfish exhibited a marked hyperplasia of pars distalis basophil (gonadotrophic) cells, which first became evident in 3-to 4-year-old fish, the age at which carp  ...  ACKNOWLEDGMENTS We are indebted to L. Lin, B. Hicks, and M. Burke for their technical help.  ... 
pmid:884668 fatcat:zi2mopk7vnejrhbni6axbxg2jy

Pose Space Surface Manipulation

Yusuke Yoshiyasu, Nobutoshi Yamazaki
2012 International Journal of Computer Games Technology  
To incorporate example deformations, we infer a pose from the handle translations/rotations and perform pose space interpolation, thereby avoiding involved nonlinear optimization.  ...  With the translational control, the user can create a model in natural poses easily. The rotational control can adjust the local pose intuitively by bending and twisting.  ...  Acknowledgment The authors would like to thank Brett Allen for the Arm and Shoulder datasets, Robert W.  ... 
doi:10.1155/2012/596953 fatcat:to7rerasnjbi7fozyjp5ardr4u

A Unified Framework for Domain Adaptive Pose Estimation [article]

Donghyun Kim, Kaihong Wang, Kate Saenko, Margrit Betke, Stan Sclaroff
2022 arXiv   pre-print
Our method outperforms existing baselines on human pose estimation by up to 4.5 percent points (pp), hand pose estimation by up to 7.4 pp, and animal pose estimation by up to 4.8 pp for dogs and 3.3 pp  ...  We propose to align representations using both input-level and output-level cues (pixels and pose labels, respectively), which facilitates the knowledge transfer from the source domain to the unlabeled  ...  Acknowledgments This work has been partially supported by NSF Award, DARPA, ONR MURI grant N00014-19-1-2571 associated with AUSMURIB000001 (to M.B.) and by NSF grant 1551572 (to. M.B.).  ... 
arXiv:2204.00172v2 fatcat:bz7ev3xfdvdb3nmdk4bfzuvrpm

Image Animation with Keypoint Mask [article]

Or Toledano, Yanir Marmor, Dov Gertz
2021 arXiv   pre-print
We chose to tackle the generic, unsupervised setting, where we need to apply animation to any arbitrary object, without any domain specific model for the structure of the input.  ...  Motion transfer is the task of synthesizing future video frames of a single source image according to the motion from a given driving video.  ...  Unsupervised mask) coloring matching keypoints with the same color to keypoint learning for guiding class-conditional video pre- help the module to learn a motion flow.  ... 
arXiv:2112.10457v1 fatcat:x3lkug62qbak7bt2hdndg6idrm

Coarse-to-Fine Isometric Shape Correspondence by Tracking Symmetric Flips

Y. Sahillioğlu, Y. Yemez
2013 Computer graphics forum (Print)  
To this effect, we extend our previous work which handles the dense isometric correspondence problem in the original 3D Euclidean space via coarse-to-fine combinatorial matching.  ...  We compare the resulting dense correspondence algorithm with state-of-the-art techniques over several 3D shape benchmark datasets.  ...  Dataset: TOSCA The original TOSCA dataset is explored under three classes: within animals (11 cats and 8 horses), within humans (two different males with 12 and 20 poses each and 12 females), and across  ... 
doi:10.1111/cgf.12007 fatcat:2jkyqjbff5aoblckhthzxw3v7e

Rapid Formation of Extended Processes and Engagement of Theiler's Virus-Infected Neurons by CNS-Infiltrating CD8 T Cells

Jeremiah R. McDole, Steve C. Danzer, Raymund Y.K. Pun, Yi Chen, Holly L. Johnson, Istvan Pirko, Aaron J. Johnson
2010 American Journal of Pathology  
Adoptively transferred GFP؉ CD8؉ splenocytes migrated to the brain and became 93% specific for the immunodominant virus epitope D b :VP2 121-130 .  ...  We used confocal microscopy to visualize the morphology, migration , and specific cellular interactions between adoptively transferred CD8 T cells and specific CNS cell types.  ...  Acknowledgments A special thanks to Markus Simon (Max-Planck Institute of Immunobiology, Freiburg, Germany) for providing granzyme B antibody and Moses Rodriguez (Mayo Clinic, Rochester, Minnesota) for  ... 
doi:10.2353/ajpath.2010.100231 pmid:20813972 pmcid:PMC2947278 fatcat:u4kr66n6tvgsndudyf4ccmz22q

A continuous, editable representation for deforming mesh sequences with separate signals for time, pose and shape

Thomas J. Cashman, Kai Hormann
2012 Computer graphics forum (Print)  
We also show that our representation makes it possible to design new deforming sequences simply by sketching a curve in a 2D pose space.  ...  We address these limitations with a continuous representation that extracts redundancy in three separate phases, leading to separate editable signals in time, pose and shape.  ...  Figure 7 : 7 An example of deformation transfer [BVGP09]; we transfer the animation shown in (a) to an elephant model, by replacing the 8 components {h j } m=7 0 of the shape signal but keeping the time  ... 
doi:10.1111/j.1467-8659.2012.03032.x fatcat:vgyqoy5wg5hitepr7ifdfnflru

Killer Conservation: the implications of disease on gorilla conservation

Rhiannon Schultz
2016 BioRisk  
Local people living in and around sensitive habitats also pose a potential threat of disease transmission.  ...  Some of these projects and programs focus on viewing or studying endangered wildlife, but can have negative implications on wild animal health.  ...  Acknowledgements I would like to thank Miami University and Project Dragonfly for the encouragement and guidance in publishing this paper.  ... 
doi:10.3897/biorisk.11.9941 fatcat:h5d2fxeysnc3lbyj6ipuyvvw2e

High-resolution animal tracking with integration of environmental information in aquatic systems [article]

Fritz Francisco, Paul Nuehrenberg, Alex L Jordan
2020 bioRxiv   pre-print
The application is based on low-cost cameras and does not require the animals to be confined or handled in any way.  ...  This cost-effective and open-source framework allows the analysis of animal behaviour in aquatic systems at an unprecedented resolution.  ...  almost all animals below a certain size class [56] .  ... 
doi:10.1101/2020.02.25.963926 fatcat:ojee5g4t3vh23fet34v2bmdmb4

Shape Animation with Combined Captured and Simulated Dynamics [article]

Benjamin Allain, Li Wang, Jean-Sebastien Franco, Franck Hetroy, and Edmond Boyer
2016 arXiv   pre-print
We present a novel volumetric animation generation framework to create new types of animations from raw 3D surface or point cloud sequence of captured real performances.  ...  The representation makes no human specific assumption and can be used to capture and re-simulate the actor with props or other moving scenery elements.  ...  T t ,T). (7) This imposes general proximity of poses to a sequence specific "rest" pose.  ... 
arXiv:1601.01232v1 fatcat:mcul24jicvaw5axwwaxan5afru
« Previous Showing results 1 — 15 out of 15,719 results