Daeyun Shin

PhD, UC Irvine

[Thariq @trq212]

https://t.co/hPiZr1kG7r

[Daeyun Shin @DaeyunShin]

https://t.co/BpqC9FydtH

[ICCV2021 @ICCV_2021]

To encourage reviewers to consider the carefully crafted reviewer guidelines available at: https://t.co/olqUxRzZP3
we will publish daily snippet reminders of these guidelines in this thread. #ReviewerGuidelines #ICCV2021. #Day2_of_40 days reviewing period #StartEarly
1/N

[DeepAI @DeepAI]

Predicting Camera Viewpoint Improves Cross-dataset Generalization for 3D Human Pose Estimation
https://t.co/Mnv0HNm6uu
by @allen_zhe_wang et al. including @DaeyunShin
#Estimator #Statistics

[DeepAI @DeepAI]

Domain Decluttering: Simplifying Images to Mitigate Synthetic-Real Domain Shift and Improve Depth Estimation
https://t.co/K2ptpnKEgx
by Yunhan Zhao et al. including @DaeyunShin
#Estimator #Statistics

[Daeyun Shin @DaeyunShin]

SynSin: End-to-end View Synthesis from a Single Image
by Olivia Wiles et al.
A cool novel view synthesis paper that does multi-view CNN feature transformation from a single image.
https://t.co/seLjdwgFCK

[Tomasz Malisiewicz @quantombone]

StructEdit: Learning Structural Shape Variations. Cool use of autoencoders for part-based 3D shape modeling.
https://t.co/Cx8z0vTjpS #ComputerVision

[Frank Dellaert @fdellaert]

Whoa - CVPR is exploding.

[Quoc Le @quocleix]

Want to improve accuracy and robustness of your model? Use unlabeled data!

Our new work uses self-training on unlabeled data to achieve 87.4% top-1 on ImageNet, 1% better than SOTA. Huge gains are seen on harder benchmarks (ImageNet-A, C and P).

Link: https://t.co/ZYDaef6sdp

[NVIDIA AI @NVIDIAAI]

Research efforts in #3D computer vision and #AI are on the rise. To accelerate 3D #deeplearning research, NVIDIA releases Kaolin as a PyTorch library. See how researchers use Kaolin to move 3D models into the realm of neural networks. https://t.co/7wyudMnO6o

[Daeyun Shin @DaeyunShin]

Open problems in deep learning and visual SLAM, according to @quantombone at #ICCV2019 workshop.

[Alex Schwing @alexschwing]

Congratulations (축하해요!) to students @JyotiAneja, @harsh_092 and @tanmay2099 from @ECEIllinois @IllinoisCS and colleagues @DhruvBatraDB, @HoiemDerek for amazing papers @ICCV19. Stop by and 인사하세요. #ICCV #ICCV19 #Seoul

[Tomasz Malisiewicz @quantombone]

It’s very clear that Jitendra’s dream is for the computer vision community to solve core AI problems and not to build more object detectors. Check out his list of challenges: #iccv2019

[Amir Zamir @zamir_ar]

Exactly 3 years ago we proposed https://t.co/kQjRHXbkFY to #CVPR with @ozansener. Today glad to see the @nature article on importance of negative results. “one of the worst aspects of science today: its toxic definitions of success”. https://t.co/Y91osisIPA

[Eric Florenzano @ericflo]

Sometimes I wish I could fast forward a decade. The end state of this research over time is going to be so cool (and scary, as with all tech, natch.) https://t.co/Cuacmr4isC

[Daeyun Shin @DaeyunShin]

Visited @berkeley_ai to give a talk on our work on single-view 3D scene reconstruction. had a good day and learned a lot

[Daeyun Shin @DaeyunShin]

Visited @StanfordSVL to present our work on single-view 3D scene reconstruction

[Frank Dellaert @fdellaert]

Wondering about batch norm? Read this amazing thread.

[Daeyun Shin @DaeyunShin]

@quantombone I really appreciate your interest in our work. New experiments, ICCV camera-ready paper and video are available now! https://t.co/a0gDqgHpvh

[Daeyun Shin @DaeyunShin]

Camera-ready version of our ICCV paper "3D Scene Reconstruction with Multi-layer Depth and Epipolar Transformers" is out: https://t.co/a0gDqgpODJ