๐ŸŒŒ Deep Learning/Dataset

[Dataset] People-Art ๋ฐ์ดํ„ฐ์…‹

๋ณต๋งŒ 2021. 1. 11. 14:03

People-Art ๋ฐ์ดํ„ฐ์…‹์€ ์„œ๋กœ ๋‹ค๋ฅธ style์˜ ์ด๋ฏธ์ง€์—์„œ ์‚ฌ๋žŒ์„ ์ฐพ๋Š” object detection ์šฉ ๋ฐ์ดํ„ฐ์…‹์ž…๋‹ˆ๋‹ค.

 

Dataset download: github.com/BathVisArtData/PeopleArt

Paper: arxiv.org/pdf/1610.08871v1.pdf

 

 

Dataset Description

 

์„œ๋กœ ๋‹ค๋ฅธ style์˜ ์ด๋ฏธ์ง€์— ๋Œ€ํ•ด ์ž˜ ์ธ์‹ํ•˜๋„๋ก ํ•˜๋Š” ๋ฌธ์ œ๋ฅผ Cross-depiction problem์ด๋ผ๊ณ  ์ •์˜ํ•ฉ๋‹ˆ๋‹ค.

People-Art ๋ฐ์ดํ„ฐ์…‹์€ ์‚ฌ์ง„, ๋งŒํ™” ๊ทธ๋ฆฌ๊ณ  41๊ฐœ์˜ ํ™”ํ’์œผ๋กœ ๊ตฌ์„ฑ๋œ ์ด 43๊ฐœ์˜ style๋ฅผ ๊ฐ€์ง‘๋‹ˆ๋‹ค.

41๊ฐœ์˜ ํ™”ํ’ ์ด๋ฏธ์ง€๋Š” WikiArt.org์—์„œ, ์‚ฌ์ง„์€ PASCAL VOC 2012 ๋ฐ์ดํ„ฐ์…‹์—์„œ, ๋งŒํ™” ์ด๋ฏธ์ง€๋Š” Google์—์„œ ๊ฐ€์ ธ์™”๋‹ค๊ณ  ํ•ฉ๋‹ˆ๋‹ค.

 

Class๋ฅผ ์‚ฌ๋žŒ์œผ๋กœ ํ•œ์ •์ง€์€ ์ด์œ ๋Š”, ๋‹ค์–‘ํ•œ ๋ฒ”์œ„์˜ style์—์„œ ๋นˆ๋ฒˆํ•˜๊ฒŒ ๋‚˜ํƒ€๋‚˜๋Š” ๊ฒƒ์ด ์‚ฌ๋žŒ์ด๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.

 

๊ฐ style์˜ ์˜ˆ์‹œ

 

Style์˜ ์ข…๋ฅ˜๊ฐ€ ๋งค์šฐ ๋งŽ๊ณ , ์‚ฌ๋žŒ์˜ ํฌ์ฆˆ๋„ ๋งค์šฐ ๋‹ค์–‘ํ•˜์—ฌ People-Art ๋ฐ์ดํ„ฐ์…‹์ด ๋งค์šฐ ์–ด๋ ค์šด task๋ผ๊ณ  ์–˜๊ธฐํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.

 

 

Experiment

 

 

Object detection ๋ชจ๋ธ๋กœ Fast R-CNN์„ ์ด์šฉํ–ˆ์œผ๋ฉฐ, ImageNet์œผ๋กœ pretrain์„ ์ง„ํ–‰ํ–ˆ์Šต๋‹ˆ๋‹ค. ์ดํ›„์— ๋‚˜์˜จ ๋ชจ๋ธ์ธ Faster R-CNN์„ ์ด์šฉํ•˜๋ฉด ๋” ๋†’์€ ์„ฑ๋Šฅ์ด ๋‚˜์˜ค์ง€ ์•Š์„๊นŒ ์ƒ๊ฐํ•ฉ๋‹ˆ๋‹ค.

 

์•„๋ž˜๋Š” ์‹คํ—˜ ๊ฒฐ๊ณผ์ž…๋‹ˆ๋‹ค.

 

Performance

 

Fast R-CNN ์™ธ์—๋„ DPM, YOLO๋ฅผ ์‚ฌ์šฉํ–ˆ์ง€๋งŒ ์„ฑ๋Šฅ์ด ์ข‹์ง€ ์•Š์•˜์œผ๋ฉฐ,

Fast R-CNN ์ค‘์—์„œ๋„ backbone์œผ๋กœ VGG16์„ ์‚ฌ์šฉํ–ˆ์„ ๋•Œ ๊ฐ€์žฅ ์„ฑ๋Šฅ์ด ์ข‹์•˜๋‹ค๊ณ  ํ•ฉ๋‹ˆ๋‹ค.

๋ฐ˜์‘ํ˜•