Embedding 3D models in offline physical environments

Abstract

This article introduces a novel approach for embedding 3D models in offline physical environments using quick response (QR) codes. Unlike conventional methods, we consider settings where 3D models cannot be retrieved from a remote server. Our method involves generating octree models from voxelized 3D models and storing them in QR codes using a space-efficient data structure. This allows storing 3D models that are both intelligible and purposeful on standard QR codes while addressing the major storage constraint that is present in offline situations. Furthermore, we explore 3D convolutional neural networks (CNN) and autoencoders (AE) to compress 3D models with high resolutions where using octrees alone does not suffice. To the best of our knowledge, our AE network is the first to employ octrees to further compress its encoded data. Through user-friendly desktop and mobile applications, we allow users to encode, decode and visualize 3D models in augmented reality (AR) using QR codes, thus experiment with our methods. The proposed approach enables unique applications and future research in ubiquitous computing, 3D data compression and transmission, 3D AEs, AR and Virtual Reality, low-cost autonomous robots, and 3D printing.

Application of the artificial intelligence system based on graphics and vision in ethnic tourism of subtropical grasslands

“…Jiang et al (2020) [ 22 ] presented a real-time simulation for producing high-fidelity hair animation, effectively capturing non-ductility, bending, and twisting chain mechanics, along with adhesion/repulsion and detailed real-time collision effects. Ertugrul et al (2020) [ 23 ] introduced a novel method for embedding 3D models in offline physical environments using fast response codes, considering settings for 3D models that cannot be retrieved from remote servers. These methods exhibit high practicality in large-scale scene navigation.…”

Source‐enhanced prototypical alignment for single image 3D model retrieval

“…Due to the high application values of 3D models, the management of 3D big data has always been a hot topic in academia and industry. Recently, with the surge of 3D models, how to effectively retrieve 3D models from a large‐scale dataset becomes a crucial problem 14 . In summary, the mainstream ways of 3D model retrieval can be divided into traditional 3D model retrieval and image‐based cross‐domain 3D model retrieval 15,16 …”

“…Recently, with the surge of 3D models, how to effectively retrieve 3D models from a large-scale dataset becomes a crucial problem. 14 In summary, the mainstream ways of 3D model retrieval can be divided into traditional 3D model retrieval and image-based cross-domain 3D model retrieval. 15,16…”

MVCRNet: A Semi-Supervised Multi-View Framework for Robust Animal Pose Estimation with Minimal Labeled Data

“…Animal behavioral studies play a central role in agriculture, animal husbandry, and aquaculture, where there is an acute need for accurate animal posture estimation technology. Despite significant advances in 2D [1-4, 33-36, 44] and 3D [38][39][40][41][42][43] human pose estimation , the idiosyncrasies of individual animals and the scarcity of specialized data resources hinder the direct transplantation of such technologies to the animal domain. Current mainstream approaches to animal pose estimation largely rely on cross-domain learning methods [5] and synthetic data augmentation techniques [6].…”