liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Efficient 3D Reconstruction and Real-Time Rendering of Large Environments Using Gaussian Splatting
Linköping University, Department of Science and Technology.
Linköping University, Department of Science and Technology.
2026 (English)Independent thesis Advanced level (degree of Master (One Year)), 20 credits / 30 HE creditsStudent thesisAlternative title
Effektiv 3D rekonstruktion och realtidsrendering av storskaliga miljöer med Gaussian Splatting (Swedish)
Abstract [en]

Gaussian Splatting is a technique that enables photorealistic 3D reconstructions directly from photographic data, but applying the technique to large-scale indoor environments on consumer-grade hardware introduces significant computational and memory challenges. No robust pipeline has previously been demonstrated that handles data acquisition, reconstruction and real-time rendering of a complete indoor environment at this scale using only consumer-grade hardware. This thesis, conducted in collaboration with Stadium, presents, evaluates and develops a batch-based reconstruction pipeline applied to their distribution center DC100 in Norrköping, with a total area of 58,000 square meters. An analysis of time and memory complexity identified the number of simultaneously processed images as the dominant limiting factor, motivating a division of DC100 into 66 independent sections. A dedicated image acquisition strategy was developed for large-scale indoor environments and the individually reconstructed sections were merged into a coherent model via shared camera parameters. To enable real-time rendering, post-processing optimizations were applied to reduce computational demands and a distance-based level streaming system was implemented in Unreal Engine 5. Following completed reconstruction, based on 23,722 images with a total size of 79.5 GB, the work resulted in a digital model of 3 GB that despite low FPS could be navigated in real time on a laptop with integrated graphics, while on a dedicated consumer graphics card it performed at over 200 FPS. The results demonstrate that the developed method enables large-scale reconstructions with Gaussian Splatting that can be generated and rendered in real time using only consumer-grade hardware, for both data acquisition, reconstruction and real-time rendering. Future work for the method includes improved robustness in environments with repetitive structures, integration of external position data to strengthen reconstruction quality, and further optimization to enable smooth real-time rendering on low-end consumer hardware.

Abstract [sv]

Gaussian Splatting är en teknik som möjliggör fotorealistiska 3D-rekonstruktioner direkt från fotografisk data, men att tillämpa tekniken på storskaliga inomhusmiljöer på konsumenthårdvara introducerar betydande beräknings- och minnesmässiga utmaningar. Ingen robust pipeline har tidigare demonstrerats som hanterar datainsamling, rekonstruktion och realtidsrendering av en komplett inomhusmiljö i denna skala med enbart konsumenthårdvara. Detta examensarbete, utfört i samarbete med Stadium, presenterar, utvärderar och utvecklar en batchbaserad rekonstruktionspipeline tillämpad på deras distributionscenter DC100 i Norrköping, med en total yta på 58 000 kvadratmeter. En analys av tids- och minneskomplexiteten identifierade att antalet simultant processerade bilder är den dominerande begränsande faktorn, vilket motiverade en uppdelning av DC100 i 66 oberoende sektioner. En dedikerad fotostrategi utvecklades för storskaliga inomhusmiljöer och de individuellt rekonstruerade sektionerna sammanfogades till en sammanhängande modell via delade kameraparametrar. För att möjliggöra realtidsrendering tillämpades även optimeringar i efterhand för att reducera beräkningskraven och ett distansbaserat level streaming-system implementerades i Unreal Engine 5. Efter färdig rekonstruktion, baserat på 23 722 bilder med en total storlek på 79,5 GB, resulterade arbetet i en digital modell på 3 GB som trots låg FPS kunde navigeras i realtid på en laptop med integrerat grafikkort, medan den på ett dedikerat konsumentgrafikkort presterade i över 200 FPS. Resultaten demonstrerar att den utvecklade metoden möjliggör storskaliga rekonstruktioner med Gaussian Splatting som kan genereras och renderas i realtid med enbart konsumenthårdvara, för både datainsamling, rekonstruktion och realtidsrendering. Framtida arbete för metoden inkluderar förbättrad robusthet i miljöer med repetitiva strukturer, integrering av extern positionsdata för att stärka rekonstruktionskvaliteten samt vidare optimering för att möjliggöra smidig realtidsrendering även på lågpresterande konsumenthårdvara.

Place, publisher, year, edition, pages
2026. , p. 76
Keywords [en]
Gaussian Splatting, 3D reconstruction, real-time rendering, large-scale reconstruction, Unreal Engine 5, image-based reconstruction, Structure-from-Motion, batch-based reconstruction, level streaming, point cloud
Keywords [sv]
Gaussian Splatting, 3D-rekonstruktion, realtidsrendering, storskalig rekonstruktion, Unreal Engine 5, bildbaserad rekonstruktion, Structure-from-Motion, batchbaserad rekonstruktion, punktmoln
National Category
Engineering and Technology
Identifiers
URN: urn:nbn:se:liu:diva-226066ISRN: LiU-ITN-TEK-A--26/031--SEOAI: oai:DiVA.org:liu-226066DiVA, id: diva2:2084012
External cooperation
Stadium, Norrköping
Subject / course
Media Technology
Supervisors
Examiners
Available from: 2026-07-03 Created: 2026-07-03 Last updated: 2026-07-03Bibliographically approved

Open Access in DiVA

fulltext(83618 kB)92 downloads
File information
File name FULLTEXT01.pdfFile size 83618 kBChecksum SHA-512
7b1d720ac740afa27bc722fa11a0eda8cbf8d53ac59efdd6eb7b64c87912c7bca18d19301273752056ee109a3a4cd93150c426ad0871033e2cac7e6c74f9acd4
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Hellberg, JoelPalm, Linus
By organisation
Department of Science and Technology
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 92 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 1036 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf