A5040203154- Advanced Vision and Imaging
- Computer Graphics and Visualization Techniques
- 3D Shape Modeling and Analysis
- Context-Aware Activity Recognition Systems
- Advanced Optical Imaging Technologies
- Face recognition and analysis
- Calibration and Measurement Techniques
- Spacecraft Design and Technology
- Wireless Body Area Networks
- Interactive and Immersive Displays
- Business Process Modeling and Analysis
- Spacecraft and Cryogenic Technologies
- Service-Oriented Architecture and Web Services
- Image Enhancement Techniques
- Atmospheric Ozone and Climate
- Solar Radiation and Photovoltaics
- Mobile Health and mHealth Applications
- Photocathodes and Microchannel Plates
- solar cell performance optimization
- Virtual Reality Applications and Impacts
- Planetary Science and Exploration
- IoT and Edge/Fog Computing
- Human Motion and Animation
- Solar and Space Plasma Dynamics
- 3D Surveying and Cultural Heritage
University College Lahore
2022
Occidental Petroleum (United States)
2022
Cardiff University
2021
QED Technologies (United States)
2007-2021
Bausch Health (United States)
2020
Creative Technologies (United States)
2007-2019
USC Institute for Creative Technologies
2005-2019
University of Hertfordshire
2019
University of Twente
2004-2017
Western Sydney University
2015
Novel view synthesis is a challenging and ill-posed inverse rendering problem. Neural techniques have recently achieved photorealistic image quality for this task. State-of-the-art (SOTA) neural volume approaches, however, are slow to train require minutes of inference (i.e., rendering) time high resolutions. We adopt high-capacity scene representations with periodic activations jointly optimizing an implicit surface radiance field supervised exclusively posed 2D images. Our pipeline...
We describe a set of rendering techniques for an autostereoscopic light field display able to present interactive 3D graphics multiple simultaneous viewers 360 degrees around the display. The consists high-speed video projector, spinning mirror covered by holographic diffuser, and FPGA circuitry decode specially rendered DVI signals. uses standard programmable card render over 5,000 images per second graphics, projecting 360-degree views with 1.25 degree separation up 20 updates second....
We describe a set of rendering techniques for an autostereoscopic light field display able to present interactive 3D graphics multiple simultaneous viewers 360 degrees around the display. The consists high-speed video projector, spinning mirror covered by holographic diffuser, and FPGA circuitry decode specially rendered DVI signals. uses standard programmable card render over 5,000 images per second graphics, projecting 360-degree views with 1.25 degree separation up 20 updates second....
We present a set of algorithms and an associated display system capable producing correctly rendered eye contact between three-dimensionally transmitted remote participant group observers in 3D teleconferencing system. The participant's face is scanned at 30Hz real time to autostereoscopic horizontal-parallax display, displaying him or her over more than 180° field view observable multiple observers. To render the geometry with correct perspective, we create fast vertex shader based on 6D...
We present a novel method for acquisition, modeling, compression, and synthesis of realistic facial deformations using polynomial displacement maps. Our consists an analysis phase where the relationship between motion capture markers detailed geometry is inferred, animated driven solely by sparse set markers. For analysis, we record actor wearing while performing training expression clips. real-time high-resolution deformations, including dynamic wrinkle pore detail, interleaved structured...
We present a technique for capturing the extreme dynamic range of natural illumination environments that include sun and sky, which has presented challenge traditional high photography processes. find through careful selection exposure times, aperture, neutral density filters this full can be covered in seven exposures with standard digital camera. discuss particular calibration issues such as lens vignetting, infrared sensitivity, spectral transmission must addressed. an adaptive adjustment...
No abstract available.
Remote ambulatory monitoring is widely seen as playing a key part in addressing the impending crisis health care provision. We describe two mobile solutions, one developed Netherlands and Australia. In both cases patients' biosignals are measured by means of body worn sensors which communicate wirelessly with handheld device. Alarms can be transmitted over wireless communication links to remote location, professional view via web application. The clinical purposes similar, however...
We present a process for rendering realistic facial performance with control of viewpoint and illumination. The is based on one or more high-quality geometry reflectance scans an actor in static poses, driven by video streams performance. compute optical flow correspondences between neighboring frames, sparse set frames. latter are made possible leveraging the relightability 3D to match viewpoint(s) appearance videos taken arbitrary environments. As tends proper correspondence some areas but...
We present a learning-based approach for synthesizing facial geometry at medium and fine scales from diffusely-lit texture maps. When applied to an image sequence, the synthesized detail is temporally coherent. Unlike current state-of-the-art methods [17, 5], which assume "dark deep", our model trained with measured collected using polarized gradient illumination in Light Stage [20]. This enables us produce plausible across entire face, including where previous approaches may incorrectly...
We present a multi-view stereo reconstruction technique that directly produces complete high-fidelity head model with consistent facial mesh topology. While existing techniques decouple shape estimation and tracking, our framework jointly optimizes for constraints parameterization. Our method is therefore free from drift fully parallelizable dynamic performance capture. produce highly detailed geometries artist-quality UV parameterization, including secondary elements such as eyeballs, mouth...
The location, condition, and number of the Parthenon sculptures present a considerable challenge to archeologists researchers studying this monument. Although proudly stands on Athenian Acropolis after nearly 2,500 years, many its have been damaged or lost. Since end 18th century, surviving sculptural decorations scattered museums around world. We propose strategy for digitally capturing large while minimizing impact site working under time resource constraints. Our system employs custom...
Video projectors are rapidly shrinking in size, power consumption, and cost. Such provide unprecedented flexibility to stack, arrange, aim pixels without the need for moving parts. We present a dense projector display that is optimized size resolution an autostereoscopic life-sized 3D human face with wide 110 degree field of view. Applications include teleconferencing fully synthetic characters education interactive entertainment.
Automultiscopic 3D displays allow a large number of viewers to experience content simultaneously without the hassle special glasses or head gear. Our display uses dense array video projectors generate many images with high-angular density over wide-field view. As each user moves around display, their eyes smoothly transition from one view next. The is ideal for displaying life-size human subjects as it allows natural personal interactions cues such eye-gaze and spatial hand gestures. In this...
We present a technique for achieving tracked vertical parallax multiple users using variety of autostereoscopic projector array setups, including front- and rear-projection curved display surfaces. This hybrid approach allows immediate horizontal as viewers move left right they up down, allowing cues such three-dimensional (3-D) perspective eye contact to be conveyed faithfully. use low-cost RGB-depth sensor simultaneously track viewer head positions in 3-D space, we interactively update the...
Overview In 2008, the "Digital Emily" project [Alexander et al. 2009] showed how a set of high-resolution facial expressions scanned in light stage could be rigged into real-time photoreal digital character and driven with video-based animation techniques. However, Digital Emily was rendered offline, involved just front face, never seen tight closeup. this collaboration between Activision USC ICT shown at SIGGRAPH 2013's Real-Time Live venue, we endeavoured to create real-time, human which...
Abstract We present a technique for generating microstructure‐level facial geometry by augmenting mesostructure‐level scan with detail synthesized from set of exemplar skin patches scanned at much higher resolution. Additionally, we make point‐source reflectance measurements the to characterize specular lobes this smaller scale and analyze variation both mesostructure microstructure scales. digitize polarization‐based computational illumination which considers reflection single scattering....
We present a practical framework for reproducing omnidirectional incident illumination conditions with complex spectra using light stage multispectral LED lights. For lighting acquisition, we augment standard RGB panoramic photography one or more observations of color chart numerous reflectance spectra. then solve how to drive the sources so that they best reproduce appearance charts in original lighting. Even when solving non-negative intensities, show accurate reproduction is achievable...
We present a set of algorithms and an associated display system capable producing correctly rendered eye contact between three-dimensionally transmitted remote participant group observers in 3D teleconferencing system. The participant's face is scanned at 30Hz real time to autostereoscopic horizontal-parallax display, displaying him or her over more than 180° field view observable multiple observers. To render the geometry with correct perspective, we create fast vertex shader based on 6D...