arXiv subjects

Rakesh Ranjan

Publications and source records attributed to Rakesh Ranjan.

58 records · Page 4Linked to original sources

arXiv2021

Feature-Align Network with Knowledge Distillation for Efficient Denoising

We propose an efficient neural network for RAW image denoising. Although neural network-based denoising has been extensively studied for image restoration, little attention has been given to efficient denoising for compute limited and power sensitive devices, such as smartphones and smartwatches. In this paper, we present a novel architecture and a suite of training techniques for high quality denoising in mobile devices. Our work is distinguished by three main contributions. (1) Feature-Align layer that modulates the activations of an encoder-decoder architecture with the input noisy images. The auto modulation layer enforces attention to spatially varying noise that tend to be "washed away" by successive application of convolutions and non-linearity. (2) A novel Feature Matching Loss that allows knowledge distillation from large denoising networks in the form of a perceptual content loss. (3) Empirical analysis of our efficient model trained to specialize on different noise subranges. This opens additional avenue for model size reduction by sacrificing memory for compute. Extensive experimental validation shows that our efficient model produces high quality denoising results that compete with state-of-the-art large networks, while using significantly fewer parameters and MACs. On the Darmstadt Noise Dataset benchmark, we achieve a PSNR of 48.28dB, while using 263 times fewer MACs, and 17.6 times fewer parameters than the state-of-the-art network, which achieves 49.12dB.

eess.IV↗

arXiv2020

EVRNet: Efficient Video Restoration on Edge Devices

Video transmission applications (e.g., conferencing) are gaining momentum, especially in times of global health pandemic. Video signals are transmitted over lossy channels, resulting in low-quality received signals. To restore videos on recipient edge devices in real-time, we introduce an efficient video restoration network, EVRNet. EVRNet efficiently allocates parameters inside the network using alignment, differential, and fusion modules. With extensive experiments on video restoration tasks (deblocking, denoising, and super-resolution), we demonstrate that EVRNet delivers competitive performance to existing methods with significantly fewer parameters and MACs. For example, EVRNet has 260 times fewer parameters and 958 times fewer MACs than enhanced deformable convolution-based video restoration network (EDVR) for 4 times video super-resolution while its SSIM score is 0.018 less than EDVR. We also evaluated the performance of EVRNet under multiple distortions on unseen dataset to demonstrate its ability in modeling variable-length sequences under both camera and object motion.

cs.CV↗

arXiv2020

Neutron-Irradiation Induced Magnetization and Persistent Defects at High Temperatures in Graphite

Structural as well as magnetization studies have been carried out on graphite samples irradiated by neutrons over 50 years in the CIRUS research reactor at Trombay. Neutron diffraction studies reveal that the defects in irradiated graphite samples are not well annealed and remain significant up to high temperatures much greater than 653 K where the Wigner energy is completely released. We infer that the remnant defects may be intralayer Frenkel defects, which do not store large energy, unlike the interlayer Frenkel defects that store the Wigner energy. Magnetization studies on the irradiated graphite show ferromagnetic behavior even at 300 K and a large additional paramagnetic contribution at 5 K. Ab-initio calculations based on the spin-polarized density-functional theory show that the magnetism in defected graphite is essentially confined on to a single 2-coordinated carbon atom that is located around a vacancy in the hexagonal layer.

cond-mat.mtrl-sci↗

arXiv2020

Defect topology and annihilation by cooperative cascading movement of atoms in highly neutron irradiated graphite

Graphite has been used as neutron moderator or reflector in many nuclear reactors. The irradiation of graphite in a nuclear reactor results in a complex population of defects. Heating of the irradiated graphite at high temperatures results in annihilation of the defects with release of an unusually large energy, called the Wigner energy. From various experiments on highly irradiated graphite samples from CIRUS reactor at Trombay and ab-initio simulations, we have for the first time identified various 2-, 3- and 4-coordinated topological structures in defected graphite, and provided microscopic mechanism of defect annihilation on heating and release of the Wigner energy. The annihilation process involves cascading cooperative movement of atoms in two steps involving an intermediate structure. Our work provides new insights in understanding of the defect topologies and annihilation in graphite which is of considerable importance to wider areas of graphitic materials including graphene and carbon nanotubes.

cond-mat.mtrl-sci↗