# ACTNET: end-to-end learning of feature activations and multi-stream aggregation for effective instance image retrieval

arXiv: https://arxiv.org/abs/1907.05794 · HTML: https://mlsanity.com/paper/1907.05794/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Jul 2019.

## Built on by (most cited first, 7 of 7)

- [DETRs with Collaborative Hybrid Assignments Training](https://mlsanity.com/paper/2211.12860/) (2022)
- [Globally Correlation-Aware Hard Negative Generation](https://mlsanity.com/paper/2411.13145/) (2024)
- [Single-cell Subcellular Protein Localisation Using Novel Ensembles of Diverse Deep Architectures](https://mlsanity.com/paper/2205.09841/) (2022)
- [RLM-Tracking: Online Multi-Pedestrian Tracking Supported by Relative Location Mapping](https://mlsanity.com/paper/2210.10477/) (2022)
- [Exploring the Relationship: Transformative Adaptive Activation Functions in Comparison to Other Activation Functions](https://mlsanity.com/paper/2402.09249/) (2024)
- [Understanding the Distributions of Aggregation Layers in Deep Neural Networks](https://mlsanity.com/paper/2107.04458/) (2021)
- [KnifeHunter: Structured Local Representation Learning for Fine-Grained Knife Image Retrieval in Law Enforcement](https://mlsanity.com/paper/2608.07057/) (2026)

## Abstract (arXiv)

We propose a novel CNN architecture called ACTNET for robust instance image retrieval from large-scale datasets. Our key innovation is a learnable activation layer designed to improve the signal-to-noise ratio (SNR) of deep convolutional feature maps. Further, we introduce a controlled multi-stream aggregation, where complementary deep features from different convolutional layers are optimally transformed and balanced using our novel activation layers, before aggregation into a global descriptor. Importantly, the learnable parameters of our activation blocks are explicitly trained, together with the CNN parameters, in an end-to-end manner minimising triplet loss. This means that our network jointly learns the CNN filters and their optimal activation and aggregation for retrieval tasks. To our knowledge, this is the first time parametric functions have been used to control and learn optimal aggregation. We conduct an in-depth experimental study on three non-linear activation functions: Sine-Hyperbolic, Exponential and modified Weibull, showing that while all bring significant gains the Weibull function performs best thanks to its ability to equalise strong activations. The results clearly demonstrate that our ACTNET architecture significantly enhances the discriminative power of deep features, improving significantly over the state-of-the-art retrieval results on all datasets.