Real-time 3D model acquisition

Szymon Rusinkiewicz, Olaf Hall-Holt, Marc Levoy
<span title="">2002</span> <i title="ACM Press"> <a target="_blank" rel="noopener" href="" style="color: black;">ACM Transactions on Graphics</a> </i> &nbsp;
The digitization of the 3D shape of real objects is a rapidly expanding field, with applications in entertainment, design, and archaeology. We propose a new 3D model acquisition system that permits the user to rotate an object by hand and see a continuously-updated model as the object is scanned. This tight feedback loop allows the user to find and fill holes in the model in real time, and determine when the object has been completely covered. Our system is based on a 60 Hz. structured-light
more &raquo; ... gefinder, a real-time variant of ICP (iterative closest points) for alignment, and point-based merging and rendering algorithms. We demonstrate the ability of our prototype to scan objects faster and with greater ease than conventional model acquisition pipelines. Hardware: The system uses a Compaq MP1800 DLP projector, with a maximum resolution of 1024x768. Because of the need to synchronize it with the video camera, we currently send an S-Video signal to the projector, limiting us to a resolution of 640x240 interlaced. The camera we use is a Sony DXC-LS1 NTSC camera, with a 1/500 sec. shutter speed. The video is digitized by a Pinnacle Studio DC10+ capture card, yielding interlaced 640x240 video fields at 60 Hz. CPU Usage: Our prototype uses a dual-CPU system, with Intel Pentium III Xeon processors running at 1 GHz. One CPU is used for the first few stages of the range scanning pipeline, namely grabbing video frames, finding stripe boundaries, matching the boundaries across time, and identifying the boundaries from the accumulated illumination history. The second CPU performs triangulation to find 3D points, aligns the scans using the fast ICP algorithm, integrates range images into the 3D grid, and renders the updated grid. The first piece of this pipeline operates at full speed (60 Hz.), while the second operates slower, approximately 10 Hz. The reason for choosing this unequal division of stages among CPUs is to ensure that the matching stage does not drop frames; this permits the highest-possible speeds for object motion. It is not as critical for the rest of the pipeline to run at the full 60 Hz. camera rate, since this only results in a lower frame rate for the display. Layout: The layout of the system determines its working volume and resolution. For the scans presented here, we have positioned the camera and projector 20 cm. apart, with a triangulation angle of 21 degrees. This configuration produces a working volume approximately 10 cm. across. Near the front of the working volume, samples are spaced roughly every 0.5 mm. in Y (parallel to the stripe direction) and every 0.75 mm. in X (perpendicular to the stripes).
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="">doi:10.1145/566570.566600</a> <a target="_blank" rel="external noopener" href="">fatcat:cehjyt33avdohh73g6kwmltoqa</a> </span>
<a target="_blank" rel="noopener" href="" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href=""> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> </button> </a>