- Speech and Audio Processing
- Hearing Loss and Rehabilitation
- Color Science and Applications
- Advanced Adaptive Filtering Techniques
- Image and Signal Denoising Methods
- Acoustic Wave Phenomena Research
- Advanced Data Compression Techniques
- Industrial Vision Systems and Defect Detection
- Archaeological Research and Protection
- Image Enhancement Techniques
- 3D Surveying and Cultural Heritage
- Noise Effects and Management
- Music and Audio Processing
- Remote Sensing and LiDAR Applications
- Digital Media Forensic Detection
- Visual perception and processing mechanisms
- Image Processing and 3D Reconstruction
- Color perception and design
- Advanced Steganography and Watermarking Techniques
- Advancements in Photolithography Techniques
- Human auditory perception and evaluation
- Computer Graphics and Visualization Techniques
- Medical Image Segmentation Techniques
Princeton University
2022-2025
Applied Research in Acoustics (United States)
2024
Dalian Polytechnic University
2021
Chinese Academy of Cultural Heritage
2020
Peking University
2018-2019
IBM (United States)
1998
Abstract. Laser scanning or photogrammetry are useful individual techniques for digital documentation of cultural heritage sites. However, these limited usage if such as the Great Wall is in harsh geographical conditions. The usually built on ridge with cliffs both sides, so it very difficult to construct scaffolding. Therefore, three-dimensional (3D) data obtained from traditional 3D laser not complete. As UAV cannot enter enemy tower, structure inside tower unmanned aerial vehicle (UAV)...
DATA REPORT article Front. Signal Process., 29 April 2024Sec. Audio and Acoustic Processing Volume 4 - 2024 | https://doi.org/10.3389/frsip.2024.1380060
The basic Ambisonics decoding method will break down when the playback loudspeakers distribute unevenly. This paper proposes a modified method, matching projection for solving this problem. is kind of greedy algorithm. It firstly calculates value object signal over each loudspeakers, then maximum assigned to corresponding loudspeaker. process repeated until all have been gain value. objective and subjective experiments were performed evaluate proposed system system. Objective evaluation...
In this paper, a method for modeling distance dependent head-related transfer functions is presented. The HRTFs are first decomposed by spatial principal component analysis. Using deep neural networks, we model the weights of different distances. Then realize prediction in arbitrary objective and subjective experiments conducted to evaluate proposed variation function model, results have shown that has less spectral distortions than virtual sound generated better performance terms localization.
Two isolation performance metrics, inter-zone (IZI) and inter-program (IPI), are introduced for evaluating personal sound zone (PSZ) systems. Compared to the commonly used acoustic contrast metric, IZI IPI generalized multichannel audio quantify of zones programs, respectively. The two metrics shown be generally non-interchangeable suitable different scenarios, such as generating dark or minimizing audio-on-audio interference (IPI). Furthermore, examples with free-field simulations presented...
Data report for the 3D3A Lab Binaural Room Impulse Response (BRIR) Dataset (https://doi.org/10.34770/6gc9-5787).
The spatial sampling of binaural room transfer functions that vary with listener movements, as required for rendering personal sound zone (PSZ) head tracking, was experimentally investigated regarding its dependencies on various factors. Through measurements the in a practical PSZ system either translational or rotational movements one two mannequin listeners, filters were generated along measurement grid and then spatially downsampled to different resolutions, at which isolation performance...
Spatial audio formats like Ambisonics are playback device layout-agnostic and well-suited for applications such as teleconferencing virtual reality. Conventional Ambisonic encoding methods often rely on spherical microphone arrays efficient sound field capture, which limits their flexibility in practical scenarios. We propose a deep learning (DL)-based approach, leveraging two-stage network architecture circular array signals into second-order (SOA) multi-speaker environments. In addition,...
A deep learning framework for dynamically rendering personal sound zones (PSZs) with head tracking is presented, utilizing a spatially adaptive neural network (SANN) that inputs listeners' coordinates and outputs PSZ filter coefficients. The SANN model trained using either simulated acoustic transfer functions (ATFs) data augmentation robustness in uncertain environments or mix of measured ATFs customization under known conditions. It found augmenting room reflections the training can more...
The extent to which the performance of personal sound zone (PSZ) reproduction systems is impacted by individualization Binaural Room Transfer Functions (BRTFs) and coupling between listeners' BRTFs was investigated experimentally.Such knowledge can be valuable for deriving rules design high-performance, robust PSZ systems.The a system consisting eight frontal mid-range loudspeakers objectively evaluated with filters designed using individualized human listener generic ones measured from...
In this era of rapid development network, the infringement textile industry is becoming more and important, which seriously restricts textiles. paper, a robust digital watermarking method based on chrominance texture features images proposed. The cover image was transformed into CIE LAB color space, B channel extracted. carrier divided blocks, each block by DCT. order to better maintain visual effect robustness watermark, watermark embedded in intermediate frequency domain Then transferred...
Abstract. The virtual restoration was one of the important protection approaches for Great Wall. With help this type technology, damaged Wall can be restored to previous state at a specific time node so that tourists visit before in computer. This paper presented framework “Moon Gate” located on XinGuangWu recovered results guide researcher repair real world without secondary damage. method divided into 4 parts: (1) collection evidence based modified scale evidence; (2) fusion DIM point...
A new high-speed color printer uses image screening methods that lead to superior-quality images. The screen rulings as well the angles are varied. cyan, magenta, and yellow screens around 200 lines per inch. black starts at 212 inch 45 degrees progresses 300 zero when half pels in basic cell on. This improves details plane without introducing significant moiré patterns.Since is capable of 4 bits/color, threshold matrices used determine onset printing. actual output intensities selected...
The paper describes a nonlinear approach for constructing color conversions based on radial basis functions (RBFs). RBF is embedded in two-layer structure that uses linear transfer function the output units and hidden units. RBFs are popular interpolating scattered data as associated system of equations guaranteed to be invertible under very mild conditions locations points. In particular, do not require lie any sort regular grid. purpose using conversion improve accuracy, efficiency,...
Recently, a fast error diffusion halftoning algorithm using look-up tables (LUT) was proposed to speed up the multiplication of filter coefficients. In this letter, we propose another LUTbased which is more flexible in terms size LUT that can be used and thus allows for optimal tradeoff between halftone quality, processing speed, hardware complexity parallelizability. Furthermore, aggregate computed with different bitdepths errors. As an example, present variant Floyd-Steinberg consists two...
Two optimization approaches are proposed to enhance the performance of personal sound zone (PSZ) systems with crosstalk cancellation (XTC). The two adjust trade-off between important attributes a system: acoustic isolation and cancellation, by either modifying cost function in problem (the direct approach) or controlling amount target transfer functions indirect filter generation process. effectiveness is evaluated using metrics inter-program (IPI) XTC level, through numerical simulations...
This study experimentally evaluates the bilateral Ambisonics method for synthesizing binaural room transfer functions (BRTFs) and explores its application in generating personal sound zones (PSZs) around listeners’ ears. Bilateral is proposed improving spatial reproduction accuracy at a limited order, by shifting origin of representation from head center to two While numerical simulations have demonstrated superiority over traditional Ambisonics, little attention has been given validating...
A new technique is described for color conversions of JPEG images. For each input block component, the conversion 63 AC coefficients processed in transform domain instead spatial domain. Only DC components are transformed to and then through traditional lookup table create color-converted output block. Given converted value block, remaining directly via scaling functions that accessed a as function only term. n-dimensional space m-dimensional conversion, n component blocks m blocks. An IDCT...
A noval novel technique is described for high speed color conversions JPEG-compressed images. The conversion processed in the transform domain instead of traditional spatial domain. Only input DC coefficients multiple components are through table lookup to create output coefficients. Given each component's value, linearity its 63 AC determined and their scaling functions looked up a 1-D as function only component term. For n-dimensional space m-dimensional conversion, n blocks m blocks....
Two isolation performance metrics, Inter-Zone Isolation (IZI) and Inter-Program (IPI), are introduced for evaluating Personal Sound Zone (PSZ) systems. Compared to the commonly-used Acoustic Contrast metric, IZI IPI generalized multichannel audio, quantify of sound zones audio programs, respectively. The two metrics shown be generally non-interchangeable suitable different scenarios, such as generating dark or minimizing audio-on-audio interference (IPI). Furthermore, examples with...
In a complex printing environment, color conversions occur at many diverse places throughout the system. an Advanced Function Presentation on Postscript data, image data and text graphics. It is particularly important to produce consistent colors across all paths for both device-dependent device-independent spaces.As example, EPS file may be RIPped in Infoprint Manager (IPM) print server converted which sent printer. Or imbedded Intelligent Printer Data Stream Thus, conversion within might...