2023 4.14 - VET: Visual Error Tomography for Point Cloud Completion and High-Quality Neural Rendering/ClipID:50627 vorhergehender Clip

Die automatischen Untertitel, die mit Whisper Open AI in diesem Video-Player (und im Multistream-Video-Player) generiert werden, dienen der Bequemlichkeit und Barrierefreiheit. Es ist jedoch zu beachten, dass die Genauigkeit und Interpretation variieren können. Für mehr Informationen lesen Sie bitte die FAQs (Absatz 14)
Aufnahme Datum 2023-11-13

Kurs-Verknüpfung

FAU Visual Computing

Zugang

Frei

Sprache

Englisch

Einrichtung

Lehrstuhl für Informatik 9 (Graphische Datenverarbeitung)

Produzent

Lehrstuhl für Informatik 9 (Graphische Datenverarbeitung)

In the last few years, deep neural networks opened the doors for big advances in novel view synthesis. Many of these approaches are based on a (coarse) proxy geometry obtained by structure from motion algorithms. Small deficiencies in this proxy can be fixed by neural rendering, but larger holes or missing parts, as they commonly appear for thin structures or for glossy regions, still lead to distracting artifacts and temporal instability. In this paper, we present a novel neural-rendering-based approach to detect and fix such deficiencies. As a proxy, we use a point cloud, which allows us to easily remove outlier geometry and to fill in missing geometry without complicated topological operations.
Keys to our approach are (i) a differentiable, blending point-based renderer that can blend out redundant points, as well as (ii) the concept of Visual Error Tomography (VET), which allows us to lift 2D error maps to identify 3D-regions lacking geometry and to spawn novel points accordingly. Furthermore, (iii) by adding points as nested environment maps, our approach allows us to generate high-quality renderings of the surroundings in the same pipeline. In our results, we show that our approach can improve the quality of a point cloud obtained by structure from motion and thus increase novel view synthesis quality significantly. In contrast to point growing techniques, the approach can also fix large-scale holes and missing thin structures effectively. Rendering quality outperforms state-of-the-art methods and temporal stability is significantly improved, while rendering is possible at real-time frame rates.

Mehr Videos aus der Kategorie "Technische Fakultät"

2024-11-14
Studon
geschützte Daten  
2024-11-15
Studon
geschützte Daten  
2024-11-15
IdM-Anmeldung
geschützte Daten  
2024-11-14
Frei
freie Daten  
2024-11-15
IdM-Anmeldung
geschützte Daten