๋ณธ๋ฌธ ๋ฐ”๋กœ๊ฐ€๊ธฐ

3d generation1

[Paper Review] PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation PointNet: Deep Learning on Point Sets for 3D Classification and SegmentationCharles R. Qi* Hao Su* Kaichun Mo Leonidas J. Guibas Stanford University   AbstractPoint cloud ๋Š” ๊ธฐํ•˜ํ•™ ๋ฐ์ดํ„ฐ ๊ตฌ์กฐ์—์„œ ์ค‘์š”ํ•œ ์š”์†Œ์ด๋‹ค.๊ธฐ์กด ์—ฐ๊ตฌ์ž๋“ค์€ Point cloud ์˜ ๋น„๊ทœ์น™์ ์ธ ๊ตฌ์„ฑ ๋•Œ๋ฌธ์— ์ด๋Ÿฌํ•œ ๋ฐ์ดํ„ฐ๋ฅผ 3D voxel girds๋‚˜ Image collections๋กœ ๋ณ€ํ™˜ํ–ˆ๋‹ค.=> ์ด์™€ ๊ฐ™์€ ๊ฒฝ์šฐ ๋ฐ์ดํ„ฐ๋ฅผ ๊ณผ๋„ํ•˜๊ฒŒ ์ƒ์„ธํ•˜๊ฒŒ ์ €์žฅํ•˜๊ฑฐ๋‚˜ ์ฒ˜๋ฆฌํ•˜๋Š” ๋ฐฉ์‹์ด ๋ถˆํ•„์š”ํ•˜๊ฒŒ ๋งŽ์•„์ ธ์„œ ์„ฑ๋Šฅ์ €ํ•˜๋‚˜ ์‹œ์Šคํ…œ ๋ฌธ์ œ๊ฐ€ ๋ฐœ์ƒํ•  ์ˆ˜๋„ ์žˆ๋‹ค.๊ทธ๋ฆฌํ•˜์—ฌ, ์ด ๋…ผ๋ฌธ์—์„œ๋Š” input์œผ๋กœ ๋“ค์–ด์˜ค๋Š” points๋“ค์˜ ์ˆœ์—ด ๋ถˆ๋ณ€.. 2024. 10. 18.

ํ‹ฐ์Šคํ† ๋ฆฌํˆด๋ฐ”