Local Features and Visual Words Emerge in Activations - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Preprints, Working Papers, ... Year : 2019

Local Features and Visual Words Emerge in Activations


We propose a novel method of deep spatial matching (DSM) for image retrieval. Initial ranking is based on image descriptors extracted from convolutional neural network activations by global pooling, as in recent state-of-the-art work. However, the same sparse 3D activation tensor is also approximated by a collection of local features. These local features are then robustly matched to approximate the optimal alignment of the tensors. This happens without any network modification, additional layers or training. No local feature detection happens on the original image. No local feature descriptors and no visual vocabulary are needed throughout the whole process. We experimentally show that the proposed method achieves the state-of-the-art performance on standard benchmarks across different network architectures and different global pooling methods. The highest gain in performance is achieved when diffusion on the nearest-neighbor graph of global descriptors is initiated from spatially verified images.
Fichier principal
Vignette du fichier
1905.06358.pdf (7.76 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02370209 , version 1 (19-11-2019)



Oriane Siméoni, Yannis Avrithis, Ondřej Chum. Local Features and Visual Words Emerge in Activations. 2019. ⟨hal-02370209⟩
53 View
72 Download



Gmail Facebook X LinkedIn More