Sökning: id:"swepub:oai:lup.lub.lu.se:981ccdb2-f80d-4ce4-b744-690fe33f6107" >
Points to patches: ...
Points to patches: Enabling the use of self-attention for 3D shape recognition
-
- Berg, Axel (författare)
- Lund University,Lunds universitet,Matematik LTH,Matematikcentrum,Institutioner vid LTH,Lunds Tekniska Högskola,Mathematics (Faculty of Engineering),Centre for Mathematical Sciences,Departments at LTH,Faculty of Engineering, LTH
-
- Oskarsson, Magnus (författare)
- Lund University,Lunds universitet,Mathematical Imaging Group,Forskargrupper vid Lunds universitet,Matematik LTH,Matematikcentrum,Institutioner vid LTH,Lunds Tekniska Högskola,Lund University Research Groups,Mathematics (Faculty of Engineering),Centre for Mathematical Sciences,Departments at LTH,Faculty of Engineering, LTH
-
- O'Connor, Mark (författare)
- ARM
-
(creator_code:org_t)
- 2022
- 2022
- Engelska 7 s.
-
Ingår i: 2022 26th International Conference on Pattern Recognition (ICPR). - 1051-4651 .- 2831-7475. - 9781665490627 - 9781665490627 ; , s. 528-534
- Relaterad länk:
-
http://dx.doi.org/10... (free)
-
visa fler...
-
https://lup.lub.lu.s...
-
https://doi.org/10.4...
-
visa färre...
Abstract
Ämnesord
Stäng
- While the Transformer architecture has become ubiquitous in the machine learning field, its adaptation to 3D shape recognition is non-trivial. Due to its quadratic computational complexity, the self-attention operator quickly becomes inefficient as the set of input points grows larger. Furthermore, we find that the attention mechanism struggles to find useful connections between individual points on a global scale. In order to alleviate these problems, we propose a two-stage Point Transformer-in-Transformer (Point-TnT) approach which combines local and global attention mechanisms, enabling both individual points and patches of points to attend to each other effectively. Experiments on shape classification show that such an approach provides more useful features for downstream tasks than the baseline Transformer, while also being more computationally efficient. In addition, we also extend our method to feature matching for scene reconstruction, showing that it can be used in conjunction with existing scene reconstruction pipelines.
Ämnesord
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
Publikations- och innehållstyp
- kon (ämneskategori)
- ref (ämneskategori)
Hitta via bibliotek
Till lärosätets databas