Skip to search formSkip to main contentSkip to account menu

MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications

This work introduces two simple global hyper-parameters that efficiently trade off between latency and accuracy and demonstrates the effectiveness of MobileNets across a wide range of applications and use cases including object detection, finegrain classification, face attributes and large scale geo-localization.
View PDF on arXiv (opens in a new tab)

FaceNet: A unified embedding for face recognition and clustering

A system that directly learns a mapping from face images to a compact Euclidean space where distances directly correspond to a measure offace similarity, and achieves state-of-the-art face recognition performance using only 128-bytes perface.
View on IEEE (opens in a new tab)

Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference

A quantization scheme is proposed that allows inference to be carried out using integer- only arithmetic, which can be implemented more efficiently than floating point inference on commonly available integer-only hardware.
View on IEEE (opens in a new tab)

MnasFPN: Learning Latency-Aware Pyramid Architecture for Object Detection on Mobile Devices

This work proposes MnasFPN, a mobile-friendly search space for the detection head, and combines it with latency-aware architecture search to produce efficient object detection models.
View on IEEE (opens in a new tab)

Looking Fast and Slow: Memory-Guided Mobile Video Object Detection

This paper addresses the analogous question of whether using memory in computer vision systems can not only improve the accuracy of object detection in video streams, but also reduce the computation time by interleaving conventional feature extractors with extremely lightweight ones which only need to recognize the gist of the scene.
View PDF on arXiv (opens in a new tab)

[Re] Explaining in Style: Training a GAN to explain a classifier in StyleSpace

StylEx is domain and classifier-agnostic, while its explanations are claimed to be human5 interpretable, distinct, coherent and sufficient to produce flipped classifier decisions.

Supplementary Materials for AQD: Towards Accurate Quantized Object Detection

From the results, the proposed method outperforms other methods by a large margin, and achieves 2.7% and 3.5% higher Top-1 accuracy for 4-bit MobileNetV1 and mobileNetV2.

O-Nect: Open Source Interface for Motion Capture using RGB Camera

Complex movement animations and realistic physical interactions can be easily recreated using this approach with minimal hardware investment.
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)