# ASMR: Learning Attribute-Based Person Search with Adaptive Semantic Margin Regularizer

arXiv: https://arxiv.org/abs/2108.04533 · HTML: https://mlsanity.com/paper/2108.04533/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 53 references that are in the map.
- First posted on arXiv 10 Aug 2021.

## Builds on (most cited first, 29 of 29)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [CosFace: Large Margin Cosine Loss for Deep Face Recognition](https://mlsanity.com/paper/1801.09414/) (2018)
- [SphereFace: Deep Hypersphere Embedding for Face Recognition](https://mlsanity.com/paper/1704.08063/) (2017)
- [Beyond Part Models: Person Retrieval with Refined Part Pooling (and a Strong Convolutional Baseline)](https://mlsanity.com/paper/1711.09349/) (2017)
- [Deep Metric Learning via Lifted Structured Feature Embedding](https://mlsanity.com/paper/1511.06452/) (2015)
- [Multi-Similarity Loss with General Pair Weighting for Deep Metric Learning](https://mlsanity.com/paper/1904.06627/) (2019)
- [No Fuss Distance Metric Learning using Proxies](https://mlsanity.com/paper/1703.07464/) (2017)
- [Proxy Anchor Loss for Deep Metric Learning](https://mlsanity.com/paper/2003.13911/) (2020)
- [Learning Fine-grained Image Similarity with Deep Ranking](https://mlsanity.com/paper/1404.4661/) (2014)
- [Joint Detection and Identification Feature Learning for Person Search](https://mlsanity.com/paper/1604.01850/) (2016)
- [On Deep Multi-View Representation Learning: Objectives and Optimization](https://mlsanity.com/paper/1602.01024/) (2016)
- [Person Search with Natural Language Description](https://mlsanity.com/paper/1702.05729/) (2017)
- [Learning Deep Structure-Preserving Image-Text Embeddings](https://mlsanity.com/paper/1511.06078/) (2015)
- [Human Semantic Parsing for Person Re-identification](https://mlsanity.com/paper/1804.00216/) (2018)
- [Improving Person Re-identification by Attribute and Identity Learning](https://mlsanity.com/paper/1703.07220/) (2017)
- [HydraPlus-Net: Attentive Deep Features for Pedestrian Analysis](https://mlsanity.com/paper/1709.09930/) (2017)
- [Camera Style Adaptation for Person Re-identification](https://mlsanity.com/paper/1711.10295/) (2017)
- [Learnable PINs: Cross-Modal Embeddings for Person Identity](https://mlsanity.com/paper/1805.00833/) (2018)
- [Deep Spatial Feature Reconstruction for Partial Person Re-identification: Alignment-Free Approach](https://mlsanity.com/paper/1801.00881/) (2018)
- [Learning Context Graph for Person Search](https://mlsanity.com/paper/1904.01830/) (2019)
- [Identity-Aware Textual-Visual Matching with Latent Co-attention](https://mlsanity.com/paper/1708.01988/) (2017)
- [Person Search via A Mask-Guided Two-Stream CNN Model](https://mlsanity.com/paper/1807.08107/) (2018)
- [Learning Cross-Modal Embeddings with Adversarial Networks for Cooking Recipes and Food Images](https://mlsanity.com/paper/1905.01273/) (2019)
- [Person Search by Multi-Scale Matching](https://mlsanity.com/paper/1807.08582/) (2018)
- [Neural Person Search Machines](https://mlsanity.com/paper/1707.06777/) (2017)
- [Linking Image and Text with 2-Way Nets](https://mlsanity.com/paper/1608.07973/) (2016)
- [Symbiotic Adversarial Learning for Attribute-based Person Search](https://mlsanity.com/paper/2007.09609/) (2020)
- [Maximum Margin Metric Learning Over Discriminative Nullspace for Person Re-identification](https://mlsanity.com/paper/1807.10908/) (2018)

## Built on by (most cited first, 11 of 11)

- [Improving Cross-Modal Retrieval with Set of Diverse Embeddings](https://mlsanity.com/paper/2211.16761/) (2022)
- [Multi-Prompts Learning with Cross-Modal Alignment for Attribute-based Person Re-Identification](https://mlsanity.com/paper/2312.16797/) (2023)
- [PARFormer: Transformer-based Multi-Task Network for Pedestrian Attribute Recognition](https://mlsanity.com/paper/2304.07230/) (2023)
- [SequencePAR: Understanding Pedestrian Attributes via A Sequence Generation Paradigm](https://mlsanity.com/paper/2312.01640/) (2023)
- [UPAR: Unified Pedestrian Attribute Recognition and Person Retrieval](https://mlsanity.com/paper/2209.02522/) (2022)
- [Rethinking Person Re-Identification via Semantic-Based Pretraining](https://mlsanity.com/paper/2110.05074/) (2021)
- [From Attributes to Natural Language: A Survey and Foresight on Text-based Person Re-identification](https://mlsanity.com/paper/2408.00096/) (2024)
- [VFM-Det: Towards High-Performance Vehicle Detection via Large Foundation Models](https://mlsanity.com/paper/2408.13031/) (2024)
- [CLEAR: Cross-Transformers with Pre-trained Language Model is All you need for Person Attribute Recognition and Retrieval](https://mlsanity.com/paper/2403.06119/) (2024)
- [Attribute-Aware Implicit Modality Alignment for Text Attribute Person Search](https://mlsanity.com/paper/2406.03721/) (2024)
- [ViTA-PAR: Visual and Textual Attribute Alignment with Attribute Prompting for Pedestrian Attribute Recognition](https://mlsanity.com/paper/2506.01411/) (2025)

## Abstract (arXiv)

Attribute-based person search is the task of finding person images that are best matched with a set of text attributes given as query. The main challenge of this task is the large modality gap between attributes and images. To reduce the gap, we present a new loss for learning cross-modal embeddings in the context of attribute-based person search. We regard a set of attributes as a category of people sharing the same traits. In a joint embedding space of the two modalities, our loss pulls images close to their person categories for modality alignment. More importantly, it pushes apart a pair of person categories by a margin determined adaptively by their semantic distance, where the distance metric is learned end-to-end so that the loss considers importance of each attribute when relating person categories. Our loss guided by the adaptive semantic margin leads to more discriminative and semantically well-arranged distributions of person images. As a consequence, it enables a simple embedding model to achieve state-of-the-art records on public benchmarks without bells and whistles.