# Aesthetic Image Captioning From Weakly-Labelled Photographs

arXiv: https://arxiv.org/abs/1908.11310 · HTML: https://mlsanity.com/paper/1908.11310/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Aug 2019.

## Built on by (most cited first, 12 of 12)

- [Q-Align: Teaching LMMs for Visual Scoring via Discrete Text-Defined Levels](https://mlsanity.com/paper/2312.17090/) (2023)
- [VILA: Learning Image Aesthetics from User Comments with Vision-Language Pretraining](https://mlsanity.com/paper/2303.14302/) (2023)
- [Understanding Aesthetics with Language: A Photo Critique Dataset for Aesthetic Assessment](https://mlsanity.com/paper/2206.08614/) (2022)
- [UniQA: Unified Vision-Language Pre-training for Image Quality and Aesthetic Assessment](https://mlsanity.com/paper/2406.01069/) (2024)
- [StyleBabel: Artistic Style Tagging and Captioning](https://mlsanity.com/paper/2203.05321/) (2022)
- [Aesthetically Relevant Image Captioning](https://mlsanity.com/paper/2211.15378/) (2022)
- [Image Aesthetics Assessment Using Graph Attention Network](https://mlsanity.com/paper/2206.12869/) (2022)
- [Ranking-aware adapter for text-driven image ordering with CLIP](https://mlsanity.com/paper/2412.06760/) (2024)
- [Can MLLMs Critique Like Humans? Evaluating Open-Ended Aesthetic Reasoning in Multimodal Large Language Models](https://mlsanity.com/paper/2606.29689/) (2026)
- [Ranking-aware Reinforcement Learning for Ordinal Ranking](https://mlsanity.com/paper/2601.20585/) (2026)
- [Aesthetic Attributes Assessment of Images with AMANv2 and DPC-CaptionsV2](https://mlsanity.com/paper/2208.04522/) (2022)
- [Aesthetic Image Captioning with Saliency Enhanced MLLMs](https://mlsanity.com/paper/2509.04378/) (2025)

## Abstract (arXiv)

Aesthetic image captioning (AIC) refers to the multi-modal task of generating critical textual feedbacks for photographs. While in natural image captioning (NIC), deep models are trained in an end-to-end manner using large curated datasets such as MS-COCO, no such large-scale, clean dataset exists for AIC. Towards this goal, we propose an automatic cleaning strategy to create a benchmarking AIC dataset, by exploiting the images and noisy comments easily available from photography websites. We propose a probabilistic caption-filtering method for cleaning the noisy web-data, and compile a large-scale, clean dataset "AVA-Captions", (230, 000 images with 5 captions per image). Additionally, by exploiting the latent associations between aesthetic attributes, we propose a strategy for training the convolutional neural network (CNN) based visual feature extractor, the first component of the AIC framework. The strategy is weakly supervised and can be effectively used to learn rich aesthetic representations, without requiring expensive ground-truth annotations. We finally show-case a thorough analysis of the proposed contributions using automatic metrics and subjective evaluations.