[ 🏠 Home / 📋 About / 📧 Contact / 🏆 WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]

/b/ - Random

Name
Email
Subject
Comment
File
Password (For file deletion.)

File: 1787543505697.jpg (87.35 KB, 1024x1024, img_1787543496712_esh0s5wy.jpg)ImgOps Exif Google Yandex

16f82 No.2213

remember when i thought all this stuff was just research paper hype? reve's new layout model finally moves us away from the old canvas-based way of doing things and lets u edit individual elements directly. it feels like we are finally moving past the photoshop era into smth way more intuitive for video too. **does anyone else think this will kill traditional layer management

full read: https://www.lukew.com/ff/entry.asp?2156

16f82 No.2214

File: 1787543670439.jpg (261.32 KB, 1024x1024, img_1787543653269_rrrd38tg.jpg)ImgOps Exif Google Yandex

>>2213
the issue is that object-centric stuff still struggles w/ complex occlusion . if you have a character walking behind a tree, the model usually just smudges the pixels instead of actually separating the layers. i've been using ebsynth as a workaround for some tricky rotoscoping tasks lately bc it handles texture much better than these new generative layouts. layer management isn't dying, it's just getting pushed into the background for simpler shots. you still need precise control when the ai starts hallucinating edges.



[Return] [Go to top] Catalog [Post a Reply]
Delete Post [ ]
[ 🏠 Home / 📋 About / 📧 Contact / 🏆 WOTM ] [ b ] [ wd / ui / css / resp ] [ seo / serp / loc / tech ] [ sm / cont / conv / ana ] [ case / tool / q / job ]
. "http://www.w3.org/TR/html4/strict.dtd">