no code implementations • 4 Jan 2023 • Aashis Khanal, Rolando Estrada
This paper is a follow-up paper on vessel topology estimation and extraction, we use the extracted topology to perform A-V state-of-the-art Artery-Vein classification, AV ratio calculation, and vessel tortuosity measurement, all fully automated.
no code implementations • 30 Dec 2022 • Jaya Krishna Mandivarapu, Blake Camp, Rolando Estrada
In this paper, we unify these two families of approaches from the angle of active learning using self-supervised learning mainfold and propose Deep Active Learning using BarlowTwins(DALBT), an active learning method for all the datasets using combination of classifier trained along with self-supervised loss framework of Barlow Twins to a setting where the model can encode the invariance of artificially created distortions, e. g. rotation, solarization, cropping etc.
no code implementations • 4 Feb 2022 • Aashis Khanal, Saeid Motevali, Rolando Estrada
We also performed several ablation studies to separately verify the importance of the segmentation and AV labeling steps of our proposed method.
no code implementations • 1 Oct 2021 • Saeid Motevali, Aashis Khanal, Rolando Estrada
The optic disc is a crucial diagnostic feature in the eye since changes to its physiognomy is correlated with the severity of various ocular and cardiovascular diseases.
1 code implementation • 23 Jul 2021 • Mehdi Mousavi, Rolando Estrada
In particular, these synthetic datasets omit features such as refraction, dispersion and caustics due to limitations in the rendering pipeline.
Ranked #1 on Caustics Segmentation on SuperCaustics
no code implementations • 13 Nov 2020 • Blake Camp, Jaya Krishna Mandivarapu, Rolando Estrada
We demonstrate that it is possible to meta-learn a single parameter vector, which we dub a neuronal phenotype, shared by all DANs in the network, which facilitates a meta-objective during deployment.
1 code implementation • 13 Jul 2020 • Mehdi Mousavi, Aashis Khanal, Rolando Estrada
With AIP, it is trivial to capture the same image under different conditions (e. g., fidelity, lighting, etc.)
Ranked #1 on Depth Estimation on DIODE
1 code implementation • 4 Jul 2020 • Jaya Krishna Mandivarapu, Blake Camp, Rolando Estrada
The goal of active learning is to infer the informativeness of unlabeled samples so as to minimize the number of requests to the oracle.
4 code implementations • 19 Mar 2019 • Aashis Khanal, Rolando Estrada
To address this limitation, we propose a novel, stochastic training scheme for deep neural networks that better classifies the faint, ambiguous regions of the image.
1 code implementation • 25 May 2018 • Blake Camp, Jaya Krishna Mandivarapu, Rolando Estrada
We demonstrate that these low-dimensional vectors can then be used to generate high-fidelity recollections of the original weights.