Adaptative Inference Cost With Convolutional Neural Mixture Models - Archive ouverte HAL Access content directly
Conference Papers Year : 2019

Adaptative Inference Cost With Convolutional Neural Mixture Models

Abstract

Despite the outstanding performance of convolutional neural networks (CNNs) for many vision tasks, the required computational cost during inference is problematic when resources are limited. In this context, we propose Convolutional Neural Mixture Models (CNMMs), a probabilistic model embedding a large number of CNNs that can be jointly trained and evaluated in an efficient manner. Within the proposed framework, we present different mechanisms to prune subsets of CNNs from the mixture, allowing to easily adapt the computational cost required for inference. Image classification and semantic segmentation experiments show that our method achieve excellent accuracy-compute trade-offs. Moreover, unlike most of previous approaches, a single CNMM provides a large range of operating points along this trade-off, without any re-training.
Fichier principal
Vignette du fichier
CNMM_ICCV2019.pdf (9.79 Mo) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-02267564 , version 1 (19-08-2019)

Identifiers

Cite

Adrià Ruiz, Jakob Verbeek. Adaptative Inference Cost With Convolutional Neural Mixture Models. ICCV 2019 - International Conference on Computer Vision, Oct 2019, Seoul, South Korea. pp.1872-1881, ⟨10.1109/ICCV.2019.00196⟩. ⟨hal-02267564⟩
127 View
85 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More