Filters








4,989 Hits in 4.3 sec

Learning Fine-Grained Motion Embedding for Landscape Animation [article]

Hongwei Xue, Bei Liu, Huan Yang, Jianlong Fu, Houqiang Li, Jiebo Luo
2021 arXiv   pre-print
To tackle this problem, we propose a model named FGLA to generate high-quality and realistic videos by learning Fine-Grained motion embedding for Landscape Animation.  ...  Motion is crucial for landscape animation as it determines how objects move in videos. Existing methods are able to generate appealing videos by learning motion from real time-lapse videos.  ...  The authors would like to thank all users for voluntarily participating our user study.  ... 
arXiv:2109.02216v2 fatcat:idwxghazbzhcnfpoufdasw5phi

Multi-animal behavioral tracking and environmental reconstruction using drones and computer vision in the wild [article]

Benjamin Koger, Adwait Deshpande, Jeffrey T. Kerby, Jacob M. Graving, Blair R. Costelloe, Iain D. Couzin
2022 bioRxiv   pre-print
in georeferenced coordinates with high spatiotemporal resolution embedded in contemporaneous 3D landscape models of the surrounding area. 4.  ...  Methods for collecting animal behavior data in natural environments, such as direct observation and bio-logging, are typically limited in spatiotemporal resolution, the number of animals that can be observed  ...  This allows the model to learn the basic, universal aspects of imagery before being fine-tuned on the user's specific dataset.  ... 
doi:10.1101/2022.06.30.498251 fatcat:a77epova6vd6rdrbvbaqjn2qye

Deep Learning for Free-Hand Sketch: A Survey [article]

Peng Xu, Timothy M. Hospedales, Qiyue Yin, Yi-Zhe Song, Tao Xiang, Liang Wang
2022 arXiv   pre-print
(iii) Promotion of future work via a discussion of bottlenecks, open problems, and potential research directions for the community.  ...  The progress of deep learning has immensely benefited free-hand sketch research and applications.  ...  Compared with other loss functions, the main advantages of triplet loss are: (i) It helps to involve more local partial orderings and annotations to learn more fine-grained embedding space.  ... 
arXiv:2001.02600v3 fatcat:lek5sivzsrat3i52lqh2eifnia

Deployment of Customized Deep Learning based Video Analytics On Surveillance Cameras [article]

Pratik Dubal, Rohan Mahadev, Suraj Kothawade, Kunal Dargan, Rishabh Iyer
2018 arXiv   pre-print
By sharing our implementation details and the experiences learned from deploying customized deep learning models for various customers, we hope that customized deep learning based video analytics is widely  ...  We demonstrate how we custom train resource constrained models and deploy them on embedded devices without significant loss in accuracy.  ...  Deep learning has dominated the landscape of computer vision for the past few years, and almost all video analytics applications can be solved with high accuracies via deep learning.  ... 
arXiv:1805.10604v2 fatcat:xzckmwvyv5hvfik35awnvwc2ji

NeuralSim: Augmenting Differentiable Simulators with Neural Networks [article]

Eric Heiden, David Millard, Erwin Coumans, Yizhou Sheng, Gaurav S. Sukhatme
2021 arXiv   pre-print
In this work, we study the augmentation of a novel differentiable rigid-body physics engine via neural networks that is able to learn nonlinear relationships between dynamic quantities and can thus learn  ...  , and present an approach for automatically discovering useful augmentations.  ...  ACKNOWLEDGMENTS We thank Carolina Parada and Ken Caluwaerts for their helpful feedback and suggestions.  ... 
arXiv:2011.04217v2 fatcat:awwqb4fzqzb6piwuxa3zim6xnm

ScienceExamCER: A High-Density Fine-Grained Science-Domain Corpus for Common Entity Recognition [article]

Hannah Smith, Zeyu Zhang, John Culnan, Peter Jansen
2019 arXiv   pre-print
Semantic class labels are drawn from a manually-constructed fine-grained typology of 601 classes generated through a data-driven analysis of 4,239 science exam questions.  ...  work we present ScienceExamCER, a densely-labeled semantic classification corpus of 133k mentions in the science exam domain where nearly all (96%) of content words have been annotated with one or more fine-grained  ...  Ma et al. (2016) show embedding models can transfer performance in zero-shot settings on fine-grained named entity classification.  ... 
arXiv:1911.10436v1 fatcat:2yasc52vvza5tfksqgdmxizrsy

Iterative Human and Automated Identification of Wildlife Images [article]

Zhongqi Miao, Ziwei Liu, Kaitlyn M. Gaynor, Meredith S. Palmer, Stella X. Yu, Wayne M. Getz
2021 arXiv   pre-print
Recently, deep learning has significantly advanced automatic wildlife recognition.  ...  These two drawbacks can be overcome through a hybrid combination of machine learning and humans in the loop.  ...  novel and fine-grained categories often possess complex intra-and inter-class distributions.  ... 
arXiv:2105.02320v2 fatcat:ja3wbtfkpbedrinyfxlgnsmuui

A Natural and Immersive Virtual Interface for the Surgical Safety Checklist Training

Andrea Ferracani, Daniele Pezzatini, Alberto Del Bimbo
2014 Proceedings of the 2014 ACM International Workshop on Serious Games - SeriousGames '14  
With the focus on natural language and entity understanding, for instance, we have improved Bing's ability to understand the user intent beyond queries and keywords.  ...  Specifically, I will talk about how we have significantly improved image search quality, and built differentiated image search user experience using NLP, entity, big data, machine learning and computer  ...  for Social Image Retrieval Affective Image Retrieval via Multi-Graph Learning Recognizing Thousands of Legal Entities through Instance-based Visual Classification Automatic Fine-grained Hyperlinking  ... 
doi:10.1145/2656719.2656725 dblp:conf/mm/FerracaniPB14a fatcat:obsb2i4iybhu3dq77hujvjtbze

Wearable performance

Johannes Birringer, Michèle Danjoux
2009 Digital Creativity  
, interactive system architecture, electronic textiles, wearable technologies, and digital animation.  ...  It then focuses on the authors' experiments with "design in motion" and digital performance, examining prototyping at the DAP-Lab which involves transdisciplinary convergences between fashion and dance  ...  Acknowledgements The "Design in Motion" project and the Suna no Onna dance environment were directed by Johannes Birringer, with fashion design and art direction by Michèle  ... 
doi:10.1080/14626260902868095 fatcat:hnhlzgf6vrgn7cq66nx33zz3ky

Oops! Predicting Unintentional Action in Video [article]

Dave Epstein, Boyuan Chen, Carl Vondrick
2019 arXiv   pre-print
We introduce a dataset of in-the-wild videos of unintentional action, as well as a suite of tasks for recognizing, localizing, and anticipating its onset.  ...  We thank NVidia for donating GPUs.  ...  Acknowledgements: We thank Dídac Surís, Parita Pooj, Hod Lipson, and Andrew McCallum for helpful discussion. Funding was provided by DARPA MCS, NSF NRI 1925157, and an Amazon Research Gift.  ... 
arXiv:1911.11206v1 fatcat:gdiefmoe55eijah5h26tgxjoym

Integrating Spatial Behavioral Ecology in Agent-Based Models for Species Conservation [chapter]

Christina A.D., Marco Musiani, Danielle J.
2011 Biodiversity  
These emergent efforts have one common, unifying feature: the need to implicitly or explicitly incorporate mechanism; that is, the underlying physiological, behavioral, and evolutionary basis for animal  ...  Such information aids in quantifying animal-habitat relationships, describing and predicting differential space use by animals, and ultimately identifying habitat that is important to an animal (Beyer  ...  Investigating habitat selection with a behavioral-ecological focus and using local, fine grain spatial parameters is common practice.  ... 
doi:10.5772/23055 fatcat:aho3dfzsjze5hp7m7lykvkdpk4

Bumblebees learn foraging routes through exploitation–exploration cycles

Jackelyn M. Kembro, Mathieu Lihoreau, Joan Garriga, Ernesto P. Raposo, Frederic Bartumeus
2019 Journal of the Royal Society Interface  
For pollinators that feed on multiple small replenishing nectar resources, the challenge is to learn efficient foraging routes while dynamically acquiring spatial information about new resource locations  ...  How animals explore and acquire knowledge from the environment is a key question in movement ecology.  ...  Radar data analysis (electronic supplementary material, §3) also reveals fine-grained behavioural changes by bumblebees when spreading out from flowers as experience progresses.  ... 
doi:10.1098/rsif.2019.0103 pmid:31288648 pmcid:PMC6685008 fatcat:ijnsjpla2nhitlp72ywdllrjbe

Multimodal Memorability: Modeling Effects of Semantics and Decay on Video Memorability [article]

Anelise Newman, Camilo Fosco, Vincent Casser, Allen Lee, Barry McNamara, Aude Oliva
2020 arXiv   pre-print
We thank Zoya Bylinskii and Phillip Isola for their useful discussions and Alex Lascelles and Mathew Monfort for helping with the dataset.  ...  This approach has the benefit that our network does not have to learn a language model, but it may not pick up on fine-grained semantic actors in the video. Captioning Results.  ...  For instance, earlier works [24, 25, 26] pointed to content of images that were predictive of their memorability (i.e. people, animals and manipulable objects are memorable, but landscapes are often  ... 
arXiv:2009.02568v1 fatcat:canniflh4vfy5ee7dt277a7g3a

Deep Learning and Synthetic Media [article]

Raphaël Millière
2022 arXiv   pre-print
After reviewing important etiological features of deep learning pipelines for media manipulation and generation, I argue that "deepfakes" and related synthetic media produced with such pipelines do not  ...  Deep learning algorithms are rapidly changing the way in which audiovisual media can be produced.  ...  Disentanglement allows them to navigate the learned distribution along human-interpretable generative factors, and thus to manipulate and generate high-quality media with fine-grained control over their  ... 
arXiv:2205.05764v1 fatcat:6th5uy6zifgydj4cbecvs3fw4u

The Autodidactic Universe [article]

Stephon Alexander, William J. Cunningham, Jaron Lanier, Lee Smolin, Stefan Stanojevic, Michael W. Toomey, Dave Wecker
2021 arXiv   pre-print
We present an approach to cosmology in which the Universe learns its own physical laws. It does so by exploring a landscape of possible laws, which we express as a certain class of matrix models.  ...  We propose that if the neural network model can be said to learn without supervision, the same can be said for the corresponding physical theory.  ...  The step of this fine-graining operation thus plays the role of time.  ... 
arXiv:2104.03902v2 fatcat:zlq7swbxa5b6xhoruxai6xr5du
« Previous Showing results 1 — 15 out of 4,989 results