Universal Sampling Rate Distortion

Boda, Vinay Praneeth; Narayan, Paresh Kumar

doi:10.1109/tit.2018.2857829

Cited by 5 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the past, many different methods have been proposed to understand the role of context [38,1,25], scale [5,37,20,26] and sampling [21,32,2,3]. Considerable importance has been given to leveraging features of different layers of the network and designing architectures for explicitly encoding context/multi-scale information [26,23,39,40] for classification.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

SNIPER: Efficient Multi-Scale Training

Singh,

Najibi,

Davis

2018

Preprint

View full text Add to dashboard Cite

We present SNIPER, an algorithm for performing efficient multi-scale training in instance level visual recognition tasks. Instead of processing every pixel in an image pyramid, SNIPER processes context regions around ground-truth instances (referred to as chips) at the appropriate scale. For background sampling, these context-regions are generated using proposals extracted from a region proposal network trained with a short learning schedule. Hence, the number of chips generated per image during training adaptively changes based on the scene complexity. SNIPER only processes 30% more pixels compared to the commonly used single scale training at 800x1333 pixels on the COCO dataset. But, it also observes samples from extreme resolutions of the image pyramid, like 1400x2000 pixels. As SNIPER operates on resampled low resolution chips (512x512 pixels), it can have a batch size as large as 20 on a single GPU even with a ResNet-101 backbone. Therefore it can benefit from batch-normalization during training without the need for synchronizing batch-normalization statistics across GPUs. SNIPER brings training of instance level recognition tasks like object detection closer to the protocol for image classification and suggests that the commonly accepted guideline that it is important to train on high resolution images for instance level visual recognition tasks might not be correct. Our implementation based on Faster-RCNN with a ResNet-101 backbone obtains an mAP of 47.6% on the COCO dataset for bounding box detection and can process 5 images per second during inference with a single GPU. Code is available at https://github.com/mahyarnajibi/SNIPER/.

show abstract

Section: Related Workmentioning

confidence: 99%

“…We use a chip size of 512 × 512 pixels. Training scales are set to (512/ms, 1.667, 3) where ms is the maximum value width and height of the image3 . The desired area ranges (i.e.…”

mentioning

confidence: 99%

SNIPER: Efficient Multi-Scale Training

Singh,

Najibi,

Davis

2018

Preprint

View full text Add to dashboard Cite

show abstract

“…Formulated as a correlated bandits problem [7], this approach seeks to identify the best arm-pair across which mutual information is minimal. Using a uniform sampling of arms, redolent of sampling mechanisms in [6], we provide an upper bound for the probability of estimation error and associated sample complexity. Our uniform sampling algorithm is similar to that in [3], [7]; however, our modified analysis takes into account estimator bias, a feature that is not common in known bandit algorithms.…”

Section: Introductionmentioning

confidence: 99%

Shared Information for a Markov Chain on a Tree

Bhattacharya

Narayan

2022

2022 IEEE International Symposium on Information Theory (ISIT)

Self Cite

View full text Add to dashboard Cite

Shared information is a measure of mutual dependence among multiple jointly distributed random variables with finite alphabets. For a Markov chain on a tree with a given joint distribution, we give a new proof of an explicit characterization of shared information. The Markov chain on a tree is shown to possess a global Markov property based on graph separation; this property plays a key role in our proofs. When the underlying joint distribution is not known, we exploit the special form of this characterization to provide a multiarmed bandit algorithm for estimating shared information, and analyze its error performance.

show abstract

Universal Single-Shot Sampling Rate Distortion

Bhattacharya

Narayan

2021

2021 IEEE International Symposium on Information Theory (ISIT)

View full text Add to dashboard Cite

Universal Sampling Rate Distortion

Cited by 5 publications

References 31 publications

SNIPER: Efficient Multi-Scale Training

SNIPER: Efficient Multi-Scale Training

Shared Information for a Markov Chain on a Tree

Universal Single-Shot Sampling Rate Distortion

Contact Info

Product

Resources

About