• 關(guān)注領(lǐng)域：

• AIGC
  

• 3D computer vision learning

• Fine-grained learning

• GNN

• 其他
  

• 聲明
  

• 論文較多，時間有限，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

Persistent Nature: A Generative Model of Unbounded 3D Worlds

https://arxiv.org/pdf/2303.13515.pdf

AIGC

Ablating Concepts in Text-to-Image Diffusion Models

https://arxiv.org/pdf/2303.13516.pdf

CMU朱俊彥團(tuán)隊的工作，解決了文本生成圖像時版權(quán)信息消除的問題．

DreamBooth3D: Subject-Driven Text-to-3D Generation

https://arxiv.org/pdf/2303.13508.pdf

不多介紹了，大神云集，爆火工作！太卷了！

CoBIT: A Contrastive Bi-directional Image-Text Generation Model

https://arxiv.org/pdf/2303.13455.pdf

UCLA的新作，一個大模型可以支持很多下游任務(wù)．不限于生成．

Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generators

https://arxiv.org/pdf/2303.13439.pdf

文本驅(qū)動的視頻生成方向，很驚訝這個領(lǐng)域發(fā)展的這么快！

TriPlaneNet: An Encoder for EG3D Inversion

https://arxiv.org/pdf/2303.13497.pdf

VLP

The effectiveness of MAE pre-pretraining for billion-scale pretraining

https://arxiv.org/pdf/2303.13496.pdf

Ross Girshick大神的工作．探討了MAE在billion-scale層面的預(yù)訓(xùn)練方法．

NS3D: Neuro-Symbolic Grounding of 3D Objects and Relations

https://arxiv.org/pdf/2303.13483.pdf