{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,7,30]],"date-time":"2025-07-30T16:05:18Z","timestamp":1753891518066,"version":"3.41.2"},"reference-count":43,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2023,11,24]],"date-time":"2023-11-24T00:00:00Z","timestamp":1700784000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Comput. Neurosci."],"abstract":"<jats:p>This paper proposes a neural network model that estimates the rotation angle of unknown objects from RGB images using an approach inspired by biological neural circuits. The proposed model embeds the understanding of rotational transformations into its architecture, in a way inspired by how rotation is represented in the ellipsoid body of <jats:italic>Drosophila<\/jats:italic>. To effectively capture the cyclic nature of rotation, the network's latent space is structured in a circular manner. The rotation operator acts as a shift in the circular latent space's units, establishing a direct correspondence between shifts in the latent space and angular rotations of the object in the world space. Our model accurately estimates the difference in rotation between two views of an object, even for categories of objects that it has never seen before. In addition, our model outperforms three state-of-the-art convolutional networks commonly used as the backbone for vision-based models in robotics.<\/jats:p>","DOI":"10.3389\/fncom.2023.1268116","type":"journal-article","created":{"date-parts":[[2023,11,24]],"date-time":"2023-11-24T14:43:45Z","timestamp":1700837025000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Bio-inspired circular latent spaces to estimate objects' rotations"],"prefix":"10.3389","volume":"17","author":[{"given":"Alice","family":"Plebe","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Mauro","family":"Da Lio","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2023,11,24]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"181855","DOI":"10.1109\/ACCESS.2020.3028740","article-title":"Object detection recognition and robot grasping based on machine learning: a survey","volume":"8","author":"Bai","year":"2020","journal-title":"IEEE Access"},{"volume-title":"Pattern Recognition and Machine Learning, Vol. 4","year":"2006","author":"Bishop","key":"B2"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1704.07911","article-title":"Explaining how a deep neural network trained with end-to-end learning steers a car","author":"Bojarski","year":"2017","journal-title":"arXiv [Preprint]"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2102.05623","article-title":"Addressing the topological defects of disentanglement via distributed operators","author":"Bouchacourt","year":"2021","journal-title":"arXiv [Preprint]"},{"key":"B5","doi-asserted-by":"publisher","first-page":"57","DOI":"10.3390\/mti2030057","article-title":"Review of deep learning methods in robotic grasp detection","volume":"2","author":"Caldera","year":"2018","journal-title":"Multimodal Technol. Interact"},{"key":"B6","doi-asserted-by":"publisher","first-page":"2961","DOI":"10.1109\/LRA.2022.3145956","article-title":"Category-association based similarity matching for novel object pick-and-place task","volume":"7","author":"Chen","year":"2022","journal-title":"IEEE Robot. Automat. Lett"},{"key":"B7","first-page":"9885","article-title":"A group-theoretic framework for data augmentation","volume":"21","author":"Chen","year":"2020","journal-title":"J. Mach. Learn. Res"},{"key":"B8","doi-asserted-by":"publisher","first-page":"149","DOI":"10.1016\/j.procir.2018.01.021","article-title":"Industrial robot control with object recognition based on deep learning","volume":"76","author":"Chen","year":"2018","journal-title":"Proc. CIRP"},{"key":"B9","doi-asserted-by":"publisher","first-page":"172","DOI":"10.1109\/TASE.2016.2600527","article-title":"Analysis and observations from the first Amazon Picking Challenge","volume":"15","author":"Correll","year":"2016","journal-title":"IEEE Transact. Automat. Sci. Eng"},{"key":"B10","doi-asserted-by":"publisher","first-page":"219","DOI":"10.1016\/0166-2236(88)90130-0","article-title":"Concurrent processing in the primate visual cortex","volume":"11","author":"DeYoe","year":"1988","journal-title":"Trends Neurosci"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1677","DOI":"10.1007\/s10462-020-09888-5","article-title":"Vision-based robotic grasping from object localization, object pose estimation to grasp estimation for parallel grippers: a review","volume":"54","author":"Du","year":"2021","journal-title":"Artif. Intell. Rev"},{"key":"B12","doi-asserted-by":"publisher","first-page":"101","DOI":"10.1038\/nature22343","article-title":"A neural circuit architecture for angular integration in Drosophila","volume":"546","author":"Green","year":"2017","journal-title":"Nature"},{"volume-title":"The Functional Organization of the Human Visual Cortex","year":"1993","author":"Gulyas","key":"B13"},{"key":"B14","doi-asserted-by":"publisher","first-page":"73","DOI":"10.1016\/j.cognition.2004.02.006","article-title":"Orientation-invariant object recognition: evidence from repetition blindness","volume":"95","author":"Harris","year":"2005","journal-title":"Cognition"},{"key":"B15","first-page":"770","article-title":"\u201cDeep residual learning for image recognition,\u201d","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"He","year":"2016"},{"key":"B16","doi-asserted-by":"publisher","first-page":"266","DOI":"10.1109\/TSMC.2020.3018325","article-title":"Deep learning in robotics: survey on model structures and training strategies","volume":"51","author":"K\u00e1roly","year":"2021","journal-title":"IEEE Transact. Syst. Man Cybernet"},{"key":"B17","first-page":"9404","article-title":"\u201cPanoptic segmentation,\u201d","volume-title":"Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition","author":"Kirillov","year":"2018"},{"key":"B18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/978-3-031-01502-1","article-title":"Deep learning for autonomous vehicle control-algorithms, state-of-the-art, and future prospects","volume":"3","author":"Kuutti","year":"2019","journal-title":"Synth. Lect. Adv. Automot. Technol"},{"key":"B19","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1038\/s41586-021-04067-0","article-title":"Building an allocentric travelling direction signal via vector computation","volume":"601","author":"Lyu","year":"2022","journal-title":"Nature"},{"key":"B20","unstructured":"Dex-Net 2.0: Deep learning to plan robust grasps with synthetic point clouds and analytic grasp metrics\n            MahlerJ.\n            LiangJ.\n            NiyazS.\n            LaskeyM.\n            DoanR.\n            LiuX.\n          10.15607\/RSS.2017.XIII.058.2017"},{"journal-title":"Columbia Object Image Library (COIL-100)","year":"1996","author":"Nene","key":"B21"},{"key":"B22","doi-asserted-by":"publisher","first-page":"765","DOI":"10.1038\/s41593-023-01308-5","article-title":"Neural dynamics and architecture of the heading direction circuit in zebrafish","volume":"26","author":"Petrucco","year":"2023","journal-title":"Nat. Neurosci"},{"key":"B23","first-page":"3406","article-title":"\u201cSupersizing self-supervision: learning to grasp from 50k tries and 700 robot hours,\u201d","volume-title":"IEEE International Conference on Robotics and Automation","author":"Pinto","year":"2016"},{"key":"B24","doi-asserted-by":"publisher","first-page":"179716","DOI":"10.1109\/ACCESS.2020.3028185","article-title":"On the road with 16 neurons: towards interpretable and manipulable latent representations for visual predictions in driving scenarios","volume":"8","author":"Plebe","year":"2020","journal-title":"IEEE Access"},{"key":"B25","first-page":"43","article-title":"\u201cMental imagery for intelligent vehicles,\u201d","volume-title":"Proceedings of the 5th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS)","author":"Plebe","year":""},{"key":"B26","first-page":"429","article-title":"\u201cDreaming mechanism for training bio-inspired driving agents,\u201d","volume-title":"Proceedings of the 2nd International Conference on Intelligent Human Systems Integration (IHSI)","author":"Plebe","year":""},{"key":"B27","doi-asserted-by":"publisher","first-page":"101169","DOI":"10.1016\/j.cogsys.2023.101169","article-title":"Human-inspired autonomous driving: a survey","volume":"83","author":"Plebe","year":"2024","journal-title":"Cogn. Syst. Res"},{"key":"B28","doi-asserted-by":"publisher","first-page":"109817","DOI":"10.1016\/j.asoc.2022.109817","article-title":"Invariance measures for neural networks","volume":"132","author":"Quiroga","year":"2023","journal-title":"Appl. Soft Comput"},{"key":"B29","first-page":"1137","article-title":"\u201cFaster R-CNN: towards real-time object detection with region proposal networks,\u201d","volume-title":"IEEE Transaction on Pattern Analysis and Machine Intelligence","author":"Ren","year":"2016"},{"key":"B30","doi-asserted-by":"crossref","first-page":"211","DOI":"10.1201\/9781351003827-8","article-title":"\u201cApplications of deep learning in robot vision,\u201d","volume-title":"Deep Learning in Computer Vision","author":"Ruiz-del-Solar","year":"2020"},{"volume-title":"Linear Representations of Finite Groups","year":"1996","author":"Serre","key":"B31"},{"key":"B32","doi-asserted-by":"publisher","first-page":"701","DOI":"10.1126\/science.171.3972.701","article-title":"Mental rotation of three-dimensional objects","volume":"171","author":"Shepard","year":"1971","journal-title":"Science"},{"key":"B33","doi-asserted-by":"crossref","first-page":"509","DOI":"10.1109\/ICRA.2014.6906903","article-title":"\u201cBigbird: a large-scale 3d database of object instances,\u201d","volume-title":"2014 IEEE International Conference on Robotics and Automation (ICRA)","author":"Singh","year":"2014"},{"key":"B34","first-page":"2818","article-title":"\u201cRethinking the Inception architecture for computer vision,\u201d","volume-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition","author":"Szegedy","year":"2016"},{"key":"B35","doi-asserted-by":"publisher","first-page":"9742","DOI":"10.1523\/JNEUROSCI.0817-07.2007","article-title":"Multimodal coding of three-dimensional rotation and translation in area MSTd: comparison of visual and vestibular selectivity","volume":"27","author":"Takahashi","year":"2007","journal-title":"J. Neurosci"},{"key":"B36","first-page":"6105","article-title":"\u201cEfficientNet: rethinking model scaling for convolutional neural networks,\u201d","volume-title":"International Conference on Machine Learning","author":"Tan","year":"2019"},{"key":"B37","doi-asserted-by":"publisher","first-page":"181","DOI":"10.1146\/annurev.neuro.29.051605.112854","article-title":"The head direction signal: origins and sensory-motor integration","volume":"30","author":"Taube","year":"2007","journal-title":"Annu. Rev. Neurosci"},{"key":"B38","doi-asserted-by":"publisher","first-page":"105842","DOI":"10.1016\/j.engappai.2023.105842","article-title":"Rotation adaptive grasping estimation network oriented to unknown objects based on novel rgb-d fusion strategy","volume":"120","author":"Tian","year":"2023","journal-title":"Eng. Appl. Artif. Intell"},{"key":"B39","doi-asserted-by":"publisher","first-page":"e23496","DOI":"10.7554\/eLife.23496","article-title":"Angular velocity integration in a fly heading circuit","volume":"6","author":"Turner-Evans","year":"2017","journal-title":"Elife"},{"key":"B40","first-page":"291","article-title":"\u201cLearning a visuomotor controller for real world robotic grasping using simulated depth images,\u201d","volume-title":"Proceedings of the 1st Annual Conference on Robot Learning","author":"Viereck","year":"2017"},{"key":"B41","doi-asserted-by":"publisher","first-page":"600","DOI":"10.1109\/TIP.2003.819861","article-title":"Image quality assessment: from error visibility to structural similarity","volume":"13","author":"Wang","year":"2004","journal-title":"IEEE Transact. Image Process"},{"key":"B42","doi-asserted-by":"publisher","first-page":"8949","DOI":"10.1016\/j.cogr.2022.03.002","article-title":"Overview of robotic grasp detection from 2D to 3D","volume":"7","author":"Yin","year":"2022","journal-title":"Cogn. Robot"},{"key":"B43","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2023.3244186.","article-title":"A novel robotic pushing and grasping method based on vision transformer and convolution","author":"Yu","year":"2023","journal-title":"IEEE Transact. Neural Netw. Learn. Syst"}],"container-title":["Frontiers in Computational Neuroscience"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fncom.2023.1268116\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,24]],"date-time":"2023-11-24T14:44:00Z","timestamp":1700837040000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fncom.2023.1268116\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,11,24]]},"references-count":43,"alternative-id":["10.3389\/fncom.2023.1268116"],"URL":"https:\/\/doi.org\/10.3389\/fncom.2023.1268116","relation":{},"ISSN":["1662-5188"],"issn-type":[{"type":"electronic","value":"1662-5188"}],"subject":[],"published":{"date-parts":[[2023,11,24]]},"article-number":"1268116"}}