Paper Reading AI Learner

Dynamic Portal Occlusion for Precomputed Interactive Sound Propagation

2021-07-24 07:04:53
Nikunj Raghuvanshi

Abstract

An immersive audio-visual experience in games and virtual reality requires fast calculation of diffraction-based acoustic effects. To maintain plausibility, the effects must retain spatial smoothness on source and listener motion within geometrically complex scenes. Precomputed wave-based techniques can render such results at low runtime CPU cost, but remain limited to static scenes. Modeling the occlusion effect of dynamic portals such as doors present an unresolved challenge to maintain audio-visual consistency. We present a fast solution implementable as a drop-in extension to existing precomputed systems. Key is a novel portal-search method that leverages precomputed propagation delay and direction data to find portals intervening the diffracted shortest path connecting dynamic source and listener at runtime. The method scales linearly with number of portals in worst case, far cheaper than explicit global path search that scales with scene area. We discuss culling techniques to accelerate further. The search algorithm is combined with geometric-acoustic approximations to model the additional direct and indirect energy loss from intervening portals depending on their dynamic closure state. We demonstrate plausible audio-visual animations within our system integrated with Unreal Engine 4 (TM) and Wwise (TM).

Abstract (translated)

URL

https://arxiv.org/abs/2107.11548

PDF

https://arxiv.org/pdf/2107.11548.pdf


Tags
3D Action Action_Localization Action_Recognition Activity Adversarial Agent Attention Autonomous Bert Boundary_Detection Caption Chat Classification CNN Compressive_Sensing Contour Contrastive_Learning Deep_Learning Denoising Detection Dialog Diffusion Drone Dynamic_Memory_Network Edge_Detection Embedding Embodied Emotion Enhancement Face Face_Detection Face_Recognition Facial_Landmark Few-Shot Gait_Recognition GAN Gaze_Estimation Gesture Gradient_Descent Handwriting Human_Parsing Image_Caption Image_Classification Image_Compression Image_Enhancement Image_Generation Image_Matting Image_Retrieval Inference Inpainting Intelligent_Chip Knowledge Knowledge_Graph Language_Model Matching Medical Memory_Networks Multi_Modal Multi_Task NAS NMT Object_Detection Object_Tracking OCR Ontology Optical_Character Optical_Flow Optimization Person_Re-identification Point_Cloud Portrait_Generation Pose Pose_Estimation Prediction QA Quantitative Quantitative_Finance Quantization Re-identification Recognition Recommendation Reconstruction Regularization Reinforcement_Learning Relation Relation_Extraction Represenation Represenation_Learning Restoration Review RNN Salient Scene_Classification Scene_Generation Scene_Parsing Scene_Text Segmentation Self-Supervised Semantic_Instance_Segmentation Semantic_Segmentation Semi_Global Semi_Supervised Sence_graph Sentiment Sentiment_Classification Sketch SLAM Sparse Speech Speech_Recognition Style_Transfer Summarization Super_Resolution Surveillance Survey Text_Classification Text_Generation Tracking Transfer_Learning Transformer Unsupervised Video_Caption Video_Classification Video_Indexing Video_Prediction Video_Retrieval Visual_Relation VQA Weakly_Supervised Zero-Shot