# Visual Concept Recognition and Localization via Iterative Introspection

arXiv: https://arxiv.org/abs/1603.04186 · HTML: https://mlsanity.com/paper/1603.04186/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Mar 2016.

## Built on by (most cited first, 6 of 6)

- [Dynamic Neural Networks: A Survey](https://mlsanity.com/paper/2102.04906/) (2021)
- [SR-GNN: Spatial Relation-aware Graph Neural Network for Fine-Grained Image Categorization](https://mlsanity.com/paper/2209.02109/) (2022)
- [Regional Attention Network (RAN) for Head Pose and Fine-grained Gesture Recognition](https://mlsanity.com/paper/2101.06634/) (2021)
- [Play It Back: Iterative Attention for Audio Recognition](https://mlsanity.com/paper/2210.11328/) (2022)
- [Switchable Representation Learning Framework with Self-compatibility](https://mlsanity.com/paper/2206.08289/) (2022)
- [Long-Distance Gesture Recognition using Dynamic Neural Networks](https://mlsanity.com/paper/2308.04643/) (2023)

## Abstract (arXiv)

Convolutional neural networks have been shown to develop internal representations, which correspond closely to semantically meaningful objects and parts, although trained solely on class labels. Class Activation Mapping (CAM) is a recent method that makes it possible to easily highlight the image regions contributing to a network's classification decision. We build upon these two developments to enable a network to re-examine informative image regions, which we term introspection. We propose a weakly-supervised iterative scheme, which shifts its center of attention to increasingly discriminative regions as it progresses, by alternating stages of classification and introspection. We evaluate our method and show its effectiveness over a range of several datasets, where we obtain competitive or state-of-the-art results: on Stanford-40 Actions, we set a new state-of the art of 81.74%. On FGVC-Aircraft and the Stanford Dogs dataset, we show consistent improvements over baselines, some of which include significantly more supervision.