Deep Quantized Representation For Enhanced Reconstruction

@article{Gupta2020DeepQR,
  title={Deep Quantized Representation For Enhanced Reconstruction},
  author={Akash Gupta and Abhishek Aich and Kevin Rodriguez and G. Venugopala Reddy and Amit K. Roy-Chowdhury},
  journal={2020 IEEE 17th International Symposium on Biomedical Imaging Workshops (ISBI Workshops)},
  year={2020},
  pages={1-4}
}
While machine learning approaches have shown remarkable performance in biomedical image analysis, most of these methods rely on high-quality and accurate imaging data. However, collecting such data requires intensive and careful manual effort. One of the major challenges in imaging the Shoot Apical Meristem (SAM) of Arabidopsis thaliana, is that the deeper slices in the z-stack suffer from different perpetual quality related problems like poor contrast and blurring. These quality related issues… 

Figures from this paper

Poisson2Sparse: Self-Supervised Poisson Denoising From a Single Image

This work explores a sparsity and dictionary learning-based approach and presents a novel self-supervised learning method for single-image denoising where the noise is approximated as a Poisson process, requiring no clean ground-truth data.

Ada-VSR: Adaptive Video Super-Resolution with Meta-Learning

Meta-learning is employed to obtain adaptive parameters, using a large-scale external dataset, that can adapt quickly to the novel condition of the given test video during the internal learning task, thereby exploiting external and internal information of a video for super-resolution.

References

SHOWING 1-9 OF 9 REFERENCES

Deep Compressed Sensing

Borrowing insights from the CS perspective, a novel way of improving GANs using gradient information from the discriminator is developed and it is shown that Generative Adversarial Nets (GANs) can be viewed as a special case in this family of models.

Neural Discrete Representation Learning

Pairing these representations with an autoregressive prior, the model can generate high quality images, videos, and speech as well as doing high quality speaker conversion and unsupervised learning of phonemes, providing further evidence of the utility of the learnt representations.

Loss Functions for Image Restoration With Neural Networks

It is shown that the quality of the results improves significantly with better loss functions, even when the network architecture is left unchanged, and a novel, differentiable error function is proposed.

Adaptive cell segmentation and tracking for volumetric confocal microscopy images of a developing plant meristem.

This work constructs an optimization function that minimizes the segmentation error, which is, in turn, estimated from the tracking results, and significantly improves both tracking and segmentation when compared to an open loop framework in which segmentation and tracking modules operate separately.

ImageJ for microscopy.

The near-comprehensive range of import filters that allow easy access to image and meta-data, a broad suite processing and analysis routine, and enthusiastic support from a friendly mailing list are invaluable for all microscopy labs and facilities-not just those on a budget.

Deep learning in neural networks: An overview

Image enhancement using a contrast measure in the compressed domain

An image enhancement algorithm for images compressed using the JPEG standard is presented, based on a contrast measure defined within the discrete cosine transform (DCT) domain that does not affect the compressibility of the original image.

Cell size and growth regulation in the Arabidopsis thaliana apical stem cell niche

The results disprove various theories for plant stem cell size/cell cycle regulation, and constitute the necessary first step in the development of integrative mechanistic theories for the coordinated regulation of cell cycle progression, cell growth, and cell size in plants.

ALANET: Adaptive Latent Attention Network for Joint Video Deblurring and Interpolation

A novel architecture, Adaptive Latent Attention Network (ALANET), is introduced, which synthesizes sharp high frame-rate videos with no prior knowledge of input frames being blurry or not, thereby performing the task of both deblurring and interpolation.