2023.03.31 ArXiv精選
關(guān)注領(lǐng)域：
AIGC
3D computer vision learning
Fine-grained learning
GNN
其他
聲明
論文較多，時(shí)間有限，本專欄無法做文章的講解，只挑選出符合PaperABC研究興趣和當(dāng)前熱點(diǎn)問題相關(guān)的論文，如果你的research topic和上述內(nèi)容有關(guān)，那本專欄可作為你的論文更新源或Paper reading list．
Paper list:
今日ArXiv共更新136篇
NeRF
NeRF-Supervised Deep Stereo
https://arxiv.org/pdf/2303.17603.pdf
AIGC
AvatarCraft: Transforming Text into Neural Human Avatars with Parameterized Shape and Pose Control
https://arxiv.org/pdf/2303.17606.pdf
基于文本的human avatar生成，引入了擴(kuò)散模型．
Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Models
https://arxiv.org/pdf/2303.17591.pdf
LayoutDiffusion: Controllable Diffusion Model for Layout-to-image Generation
https://arxiv.org/pdf/2303.17189.pdf
基于layut的圖像生成．
VLP
Beyond Appearance: a Semantic Controllable Self-Supervised Learning Framework for Human-Centric Visual Tasks
https://arxiv.org/pdf/2303.17602.pdf
阿里巴巴的工作，CVPR2023已接收．利用無標(biāo)注的human image數(shù)據(jù)，進(jìn)行自監(jiān)督預(yù)訓(xùn)練．
Going Beyond Nouns With Vision & Language Models Using Synthetic Data
https://arxiv.org/pdf/2303.17590.pdf
使用合成數(shù)據(jù)來訓(xùn)練視覺語言模型，使得其具有更加open的能力．
SoftCLIP: Softer Cross-modal Alignment Makes CLIP Stronger
https://arxiv.org/pdf/2303.17561.pdf
引入軟對(duì)比學(xué)習(xí)至CLIP.
MAE
Masked Autoencoders as Image Processors
https://arxiv.org/pdf/2303.17316.pdf
