Quality-Agnostic Image Recognition via Invertible Decoder

Kim, In-Soo; Han, Seungju; Baek, Jiwon; Park, Seong-Jin; Han, Jae-Joon; Shin, Jinwoo

doi:10.1109/cvpr46437.2021.01208

Cited by 13 publications

(17 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…TransINR [8] employs the Transformer [37] as a hypernetwork to predict latent vectors to modulate the weights of the shared MLP. In addition, Instance Pattern Composers [19] have demonstrated that modulating the weights of the second MLP layer is enough to achieve high performance of generalizable INRs. Our framework also employs the Transformer encoder, but focuses on extracting locality-aware latent features for the high performance of generalizable INR.…”

Section: Related Workmentioning

confidence: 99%

“…A generalizable INR uses a single coordinate-based MLP as a shared INR decoder F θ : R din → R dout to represent multiple data instances as a continuous function. Generalizable INR [8,11,12,19,26] extracts the R number of latent codes…”

Section: Generalizable Implicit Neural Representationmentioning

confidence: 99%

“…We remark that each previous approach employs a different number of latent codes to modulate a coordinate-based MLP. For example, a single latent vector (R = 1) is commonly extracted to modulate intermediate features of the MLP [11,12,26], while a multitude of latents (R > 1) are used to modulate its weights [8,19,35]. While we modulate the features of MLP, we extract a set of latent codes to localize the information of data to leverage the locality-awareness for latent features.…”

Section: Generalizable Implicit Neural Representationmentioning

confidence: 99%

“…to modulate the weights or intermediate features of the shared MLP model [8,11,19,35]. However, despite the advances in previous approaches, their performance is still insufficient compared with individual training of INRs per sample.…”

mentioning

confidence: 99%

“…However, previous approaches to generalizable INRs are not properly designed to leverage the locality of data entities. For example, when latent codes modulate intermediate features [11,12] or weight matrices [8,19,35] of an INR decoder, the modulation methods do not exploit a specified coordinates for decoding, which restricts the latent codes to encoding global information over all pixels without capturing local relationships between specific pixels.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Human spinal cord organoids exhibiting neural tube morphogenesis for a quantifiable drug screening system of neural tube defects

Lee

Shin

Shaker

et al. 2020

Preprint

View full text Add to dashboard Cite

The human spinal cord forms well-organized neural circuits for environment sensing and motor behavior. The three-dimensional (3D) induction of the spinal cord-like tissue from human pluripotent stem cells has been reported, but they often do not mimic morphological features of neurulation and their maturity is limited. Here, we report an advanced 3D culture system for the production of human spinal cord-like organoids (hSCOs) suitable for the scale-up and quantitative studies. The hSCOs exhibited many aspects of spinal cord development, including neurulation-like tube-forming morphogenesis, differentiation of the major spinal cord neurons and glial cells, and mature synaptic functional activities. We further demonstrated that hSCOs platform allowed quantitative and systematic high-throughput examination of the potential risk of neural tube defects induced by antiepileptic drugs. Thus, hSCOs can be used for understanding human spinal cord development, disease modeling, and toxicology screening.

show abstract

Section: Related Workmentioning

confidence: 99%