# Count What You Want: Exemplar Identification and Few-shot Counting of Human Actions in the Wild

arXiv: https://arxiv.org/abs/2312.17330 · HTML: https://mlsanity.com/paper/2312.17330/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 54 references that are in the map.
- First posted on arXiv 28 Dec 2023.

## Builds on (most cited first, 18 of 18)

- [Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting](https://mlsanity.com/paper/2012.07436/) (2020)
- [Autoformer: Decomposition Transformers with Auto-Correlation for Long-Term Series Forecasting](https://mlsanity.com/paper/2106.13008/) (2021)
- [Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition](https://mlsanity.com/paper/1804.03209/) (2018)
- [Soft-DTW: a Differentiable Loss Function for Time-Series](https://mlsanity.com/paper/1703.01541/) (2017)
- [Learning To Count Everything](https://mlsanity.com/paper/2104.08391/) (2021)
- [Class-Agnostic Counting](https://mlsanity.com/paper/1811.00472/) (2018)
- [Represent, Compare, and Learn: A Similarity-Aware Framework for Class-Agnostic Counting](https://mlsanity.com/paper/2203.08354/) (2022)
- [CounTR: Transformer-based Generalised Visual Counting](https://mlsanity.com/paper/2208.13721/) (2022)
- [Counting Out Time: Class Agnostic Video Repetition Counting in the Wild](https://mlsanity.com/paper/2006.15418/) (2020)
- [Few-shot Object Counting with Similarity-Aware Feature Enhancement](https://mlsanity.com/paper/2201.08959/) (2022)
- [Few-shot Object Counting and Detection](https://mlsanity.com/paper/2207.10988/) (2022)
- [Broadcasted Residual Learning for Efficient Keyword Spotting](https://mlsanity.com/paper/2106.04140/) (2021)
- [TransRAC: Encoding Multi-scale Temporal Correlation with Transformers for Repetitive Action Counting](https://mlsanity.com/paper/2204.01018/) (2022)
- [Context-aware and Scale-insensitive Temporal Repetition Counting](https://mlsanity.com/paper/2005.08465/) (2020)
- [Real-World Repetition Estimation by Div, Grad and Curl](https://mlsanity.com/paper/1802.09971/) (2018)
- [Crowd Counting with Decomposed Uncertainty](https://mlsanity.com/paper/1903.07427/) (2019)
- [Personalized Step Counting Using Wearable Sensors: A Domain Adapted LSTM Network Approach](https://mlsanity.com/paper/2012.08975/) (2020)
- [Viewpoint-Invariant Exercise Repetition Counting](https://mlsanity.com/paper/2107.13760/) (2021)

## Built on by (most cited first, 3 of 3)

- [What is the Right Embedding Space for Contrastive Learning in Referring Expression Counting?](https://mlsanity.com/paper/2505.22850/) (2025)
- [Decoupling What to Count and Where to See for Referring Expression Counting](https://mlsanity.com/paper/2510.24374/) (2025)
- [MoBind: Motion Binding for Fine-Grained IMU-Video Pose Alignment](https://mlsanity.com/paper/2602.19004/) (2026)

## Abstract (arXiv)

This paper addresses the task of counting human actions of interest using sensor data from wearable devices. We propose a novel exemplar-based framework, allowing users to provide exemplars of the actions they want to count by vocalizing predefined sounds ''one'', ''two'', and ''three''. Our method first localizes temporal positions of these utterances from the audio sequence. These positions serve as the basis for identifying exemplars representing the action class of interest. A similarity map is then computed between the exemplars and the entire sensor data sequence, which is further fed into a density estimation module to generate a sequence of estimated density values. Summing these density values provides the final count. To develop and evaluate our approach, we introduce a diverse and realistic dataset consisting of real-world data from 37 subjects and 50 action categories, encompassing both sensor and audio data. The experiments on this dataset demonstrate the viability of the proposed method in counting instances of actions from new classes and subjects that were not part of the training data. On average, the discrepancy between the predicted count and the ground truth value is 7.47, significantly lower than the errors of the frequency-based and transformer-based methods. Our project, code and dataset can be found at https://github.com/cvlab-stonybrook/ExRAC.