SaltiNet: scan-path prediction on 360 degree images using saliency volumes
| العنوان: | SaltiNet: scan-path prediction on 360 degree images using saliency volumes |
|---|---|
| المؤلفون: | Kevin McGuinness, Noel E. O'Connor, Xavier Giro-i-Nieto, Marc Assens |
| المساهمون: | Universitat Politècnica de Catalunya. Departament de Teoria del Senyal i Comunicacions, Universitat Politècnica de Catalunya. GPI - Grup de Processament d'Imatge i Vídeo |
| المصدر: | Recercat. Dipósit de la Recerca de Catalunya instname ICCV Workshops UPCommons. Portal del coneixement obert de la UPC Universitat Politècnica de Catalunya (UPC) |
| بيانات النشر: | IEEE Press, 2018. |
| سنة النشر: | 2018 |
| مصطلحات موضوعية: | FOS: Computer and information sciences, Artificial intelligence, scanpath, Computer science, Computer Vision and Pattern Recognition (cs.CV), Computer Science - Computer Vision and Pattern Recognition, ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION, Image processing, 02 engineering and technology, 010501 environmental sciences, 01 natural sciences, Convolutional neural network, computer vision, Neural networks (Computer science), Imatges -- Processament -- Tècniques digitals, Salience (neuroscience), Machine learning, Informàtica::Intel·ligència artificial::Representació del coneixement [Àrees temàtiques de la UPC], convolutional neural networks, 0202 electrical engineering, electronic engineering, information engineering, Xarxes neuronals (Informàtica), Artificial vision, 0105 earth and related environmental sciences, Image processing--Digital techniques, Artificial neural network, business.industry, Deep learning, Intel·ligència artificial, Visió per ordinador, saliency prediction, deep learning, Pattern recognition, Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeo [Àrees temàtiques de la UPC], Visió artificial (Robòtica), Multimedia (cs.MM), Visualization, Eye tracking, 020201 artificial intelligence & image processing, eye gaze, business, Computer Science - Multimedia |
| الوصف: | We introduce SaltiNet, a deep neural network for scanpath prediction trained on 360-degree images. The model is based on a temporal-aware novel representation of saliency information named the saliency volume. The first part of the network consists of a model trained to generate saliency volumes, whose parameters are fit by back-propagation computed from a binary cross entropy (BCE) loss over downsampled versions of the saliency volumes. Sampling strategies over these volumes are used to generate scanpaths over the 360-degree images. Our experiments show the advantages of using saliency volumes, and how they can be used for related tasks. Our source code and trained models available at https://github.com/massens/saliency-360salient-2017Test. Winner of the Best Scan-path Award at the Salient360!: Visual attention modeling for 360 degrees Images Grand Challenge of ICME 2017. Presented at the ICCV 2017 Workshop on Egocentric Perception, Interaction and Computing (EPIC) |
| وصف الملف: | application/pdf |
| اللغة: | English |
| الوصول الحر: | https://explore.openaire.eu/search/publication?articleId=doi_dedup___::05c25d1b7b2c5b685a7fbc68b3872656Test https://hdl.handle.net/2117/114891Test |
| حقوق: | OPEN |
| رقم الانضمام: | edsair.doi.dedup.....05c25d1b7b2c5b685a7fbc68b3872656 |
| قاعدة البيانات: | OpenAIRE |
| الوصف غير متاح. |