Automated reconstruction of multiple objects with individual movement based on PSP

Phase-shifting profilometry (PSP) is a workhorse 3D measurement technique in industrial inspection, but its multi-frame capture requirement means that object motion degrades reconstruction quality. When multiple objects move independently—as on a conveyor belt with separated parts—the problem is compounded: each object has a different motion trajectory, and a single global motion model cannot compensate for all of them simultaneously. This paper presents an automated multi-object PSP reconstruction pipeline. Foreground objects are detected and segmented in each fringe frame using appearance and motion cues, and individual motion trajectories are estimated per object across the fringe sequence. Motion compensation is then applied per object before phase recovery, enabling accurate 3D reconstruction of each part despite its independent movement. Crucially, the pipeline is fully automated: it requires no prior knowledge of how many objects are present or how they move, making it directly deployable on dynamic industrial scenes. Published in Optics Express (2020), the method is demonstrated on multi-object scenarios with diverse, uncorrelated motion patterns, significantly reducing motion-induced reconstruction error compared to single-object or global motion correction approaches.

Problem setting

Reconstructing multiple objects that move independently during phase-shifting profilometry (PSP) capture is challenging because standard PSP assumes a static scene: motion of any object corrupts its phase estimate and can also contaminate neighbouring objects. This paper presents an automated pipeline that detects and tracks multiple independently moving objects in the PSP fringe sequence, estimates each object motion separately, and applies per-object motion compensation before phase recovery. The result is an automated system capable of simultaneously reconstructing accurate 3D shapes of multiple moving objects without manual intervention or prior knowledge of object count.

In the broader publication record, this work appears in Optics Express, 28(19):28600–28611. The visual notes below pair the paper’s original figures with a concise reading of the method, experimental setup, and reported results.

Method and visual evidence

The method follows an optical 3D measurement pipeline: acquire coded images, recover phase or geometric cues, compensate the dominant error source, and reconstruct a reliable 3D result.

The extracted figures below show the sensing setup, algorithmic signal flow, and representative reconstruction or calibration results.

Method overview. This image is extracted from an embedded PDF image object on page 4, then recomposed for web display.

Representation and setup. This image is extracted from an embedded PDF image object on page 6, then recomposed for web display.

Experimental evidence. This image is extracted from an embedded PDF image object on page 7, then recomposed for web display.

Result comparison. This image is extracted from an embedded PDF image object on page 8, then recomposed for web display.

Additional visual result. This image is extracted from an embedded PDF image object on page 10, then recomposed for web display.

Results and impact

The evaluation reported in Optics Express, 28(19):28600–28611 uses the extracted figures above to show the method’s measurement, reconstruction, segmentation, matching, or diagnostic behavior on representative experiments. These visuals are paired with the paper’s quantitative or qualitative analysis to make the workflow easier to inspect from the homepage.

Source handling

I extracted 11 candidate image objects from paper.pdf and generated the compressed WebP figures used on this page.