Hao Tan

Hao Tan

Auto-generated profile, based on publicly available information

Hao Tan

Adobe Research

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2021 in Survey of Ophthalmology
Verified
Journal Article
Closed access
HTHao Tan
Hao Tan
PBPatricia Blasco
Patricia Blasco
TLTamorah Lewis
Tamorah Lewis
SOSusan Ostmo
Susan Ostmo
MCMichael F Chiang
Michael F Chiang
JCJohn Peter Campbell
John Peter Campbell

26 December 2022 in Current Pain and Headache Reports
Verified
Journal Article
Closed access
PDF file
HTHao Tan
Hao Tan
EYErin A. Yamamoto
Erin A. Yamamoto
MEMohammed A. Elkholy
Mohammed A. Elkholy
ARAhmed M. Raslan
Ahmed M. Raslan

6 April 2024 in Clinical Neurophysiology
Verified
Journal Article
Closed access
AOAudrey Oliger
Audrey Oliger
CNCaleb Nerison
Caleb Nerison
HTHao Tan
Hao Tan
ARAhmed Raslan
Ahmed Raslan
LELia Ernst
Lia Ernst
MKMarissa Kellogg
Marissa Kellogg

9 March 2024 in arXiv
Verified
Preprint
Open access
PDF file
YHYicong Hong
Yicong Hong
KZKai Zhang
Kai Zhang
JGJiuxiang Gu
Jiuxiang Gu
SBSai Bi
Sai Bi
YZYang Zhou
Yang Zhou
HTHao Tan
Hao Tan
We propose the first Large Reconstruction Model (LRM) that predicts the 3D model of an object from a single input image within just 5 seconds. In contrast to many previous methods that are trained on small-scale datasets such as ShapeNet in a category-specific fashion, LRM adopts a highly scalable transformer-based architecture with 500 million learnable parameters to directly predict a neural radiance field (NeRF) from the input image. We train our model in an end-to-end manner on massive multi-view data containing around 1 million objects, including both synthetic renderings from Objaverse and real captures from MVImgNet. This combination of a high-capacity model and large-scale training data empowers our model to be highly generalizable and produce high-quality 3D reconstructions from various testing inputs, including real-world in-the-wild captures and images created by generative models. Video demos and interactable 3D meshes can be found on our LRM project webpage: https://yiconghong.me/LRM.

24 October 2023 in Neurosurgery
Verified
Journal Article
Closed access
HTHao Tan
Hao Tan
SBStephen G. Bowden
Stephen G. Bowden
DSDominic A. Siler
Dominic A. Siler
BOBrannan E. O'NEill
Brannan E. O'NEill
MSMaryam N. Shahin
Maryam N. Shahin
JOJosiah N. Orina
Josiah N. Orina
BACKGROUND AND OBJECTIVES: Labeling residents as “black” or “white” clouds based on perceived or presumed workloads is a timeworn custom across medical training and practice. Previous studies examining whether such perceptions align with objective workload patterns have offered conflicting results. We assessed whether such peer-assigned labels were associated with between-resident differences in objective, on-call workload metrics in three classes of neurosurgery junior residents. In doing so, we introduce more inclusive terminology for perceived differences in workload metrics. METHODS: Residents were instructed to complete surveys to identify “sunny”, “neutral”, and “stormy” residents, reflecting least to greatest perceived workloads, of their respective classes. We retrospectively reviewed department and electronic medical records to record volume of on-call work over the first 4 months of each resident's 2nd postgraduate academic year. Inter-rater agreement of survey responses was measured using Fleiss' kappa. All statistical analyses were performed with a significance threshold of P < .05. RESULTS: Across all classes, there was strong inter-rater agreement in the identification of stormy and sunny residents (Kappa = 1.000, P = .003). While differences in on-call workload measures existed within each class, “weather” designations did not consistently reflect these differences. There were significant intraclass differences in per shift consult volume in two classes (P = .035 and P = .009); however, consult volume corresponded to a resident's weather designations in only one class. Stormy residents generally saw more emergencies and, in 2 classes, performed more bedside procedures than their peers. CONCLUSION: Significant differences in objective on-call experience exist between junior neurosurgery residents. Self- and peer-assigned weather labels did not consistently align with a pattern of these differences, suggesting that other factors contribute to such labels.

6 July 2023 in Pain Practice
Verified
Journal Article
Closed access
HTHao Tan
Hao Tan
MEMohamed A. Elkholy
Mohamed A. Elkholy
ARAhmed M. Raslan
Ahmed M. Raslan