Browse > Methodology > Quantization

Quantization

115 papers with code · Methodology

State-of-the-art leaderboards

Trend Dataset Best Method Paper title Paper Code Compare

Greatest papers with code

IR-Net: Forward and Backward Information Retention for Highly Accurate Binary Neural Networks

24 Sep 2019JDAI-CV/dabnn

Weight and activation binarization is an effective approach to deep neural network compression and can accelerate the inference by leveraging bitwise operations.

NEURAL NETWORK COMPRESSION QUANTIZATION

And the Bit Goes Down: Revisiting the Quantization of Neural Networks

12 Jul 2019facebookresearch/kill-the-bits

In this paper, we address the problem of reducing the memory footprint of convolutional network architectures.

OBJECT CLASSIFICATION QUANTIZATION

Detection of Structural Change in Geographic Regions of Interest by Self Organized Mapping: Las Vegas City and Lake Mead across the Years

29 Mar 2018JustGlowing/minisom

Time-series of satellite images may reveal important data about changes in environmental conditions and natural or urban landscape structures that are of potential interest to citizens, historians, or policymakers.

QUANTIZATION TIME SERIES

Rethinking floating point for deep learning

1 Nov 2018facebookresearch/deepfloat

In 16 bits, our log float multiply-add is 0. 59x the power and 0. 68x the area of IEEE 754 float16 fused multiply-add, maintaining the same signficand precision and dynamic range, proving useful for training ASICs as well.

QUANTIZATION

Integral Human Pose Regression

ECCV 2018 JimmySuen/integral-human-pose

State-of-the-art human pose estimation methods are based on heat map representation.

3D POSE ESTIMATION QUANTIZATION

Deep Triplet Quantization

1 Feb 2019thulab/DeepHash

We propose Deep Triplet Quantization (DTQ), a novel approach to learning deep quantization models from the similarity triplets.

IMAGE RETRIEVAL QUANTIZATION

Spreading vectors for similarity search

ICLR 2019 facebookresearch/spreadingvectors

Discretizing multi-dimensional data distributions is a fundamental step of modern indexing methods.

QUANTIZATION

Scalable Methods for 8-bit Training of Neural Networks

NeurIPS 2018 eladhoffer/convNet.pytorch

Armed with this knowledge, we quantize the model parameters, activations and layer gradients to 8-bit, leaving at a higher precision only the final step in the computation of the weight gradients.

QUANTIZATION

StereoNet: Guided Hierarchical Refinement for Real-Time Edge-Aware Depth Prediction

ECCV 2018 meteorshowers/StereoNet

A first estimate of the disparity is computed in a very low resolution cost volume, then hierarchically the model re-introduces high-frequency details through a learned upsampling function that uses compact pixel-to-pixel refinement networks.

DEPTH ESTIMATION QUANTIZATION STEREO MATCHING STEREO MATCHING HAND