# Blind Users Accessing Their Training Images in Teachable Object Recognizers

arXiv: https://arxiv.org/abs/2208.07968 · HTML: https://mlsanity.com/paper/2208.07968/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 86 references that are in the map.
- First posted on arXiv 16 Aug 2022.

## Builds on (most cited first, 17 of 17)

- [YOLOv3: An Incremental Improvement](https://mlsanity.com/paper/1804.02767/) (2018)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Matching Networks for One Shot Learning](https://mlsanity.com/paper/1606.04080/) (2016)
- [Meta-Transfer Learning for Few-Shot Learning](https://mlsanity.com/paper/1812.02391/) (2018)
- [Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models](https://mlsanity.com/paper/1708.08296/) (2017)
- [Sign Language Recognition, Generation, and Translation: An Interdisciplinary Perspective](https://mlsanity.com/paper/1908.08597/) (2019)
- [An Overview of Machine Teaching](https://mlsanity.com/paper/1801.05927/) (2018)
- [TaskNorm: Rethinking Batch Normalization for Meta-Learning](https://mlsanity.com/paper/2003.03284/) (2020)
- [Meta-DETR: Image-Level Few-Shot Object Detection with Inter-Class Correlation Exploitation](https://mlsanity.com/paper/2103.11731/) (2021)
- [Machine Teaching: A New Paradigm for Building Machine Learning Systems](https://mlsanity.com/paper/1707.06742/) (2017)
- [ORBIT: A Real-World Few-Shot Dataset for Teachable Object Recognition](https://mlsanity.com/paper/2104.03841/) (2021)
- [Discriminative k-shot learning using probabilistic models](https://mlsanity.com/paper/1706.00326/) (2017)
- [Crowdsourcing the Perception of Machine Teaching](https://mlsanity.com/paper/2002.01618/) (2020)
- [Pedestrian Detection with Wearable Cameras for the Blind: A Two-way Perspective](https://mlsanity.com/paper/2003.12122/) (2020)
- [Exploring Machine Teaching with Children](https://mlsanity.com/paper/2109.11434/) (2021)
- [AI from concrete to abstract: demystifying artificial intelligence to the general public](https://mlsanity.com/paper/2006.04013/) (2020)
- [From the Lab to People's Home: Lessons from Accessing Blind Participants' Interactions via Smart Glasses in Remote Studies](https://mlsanity.com/paper/2203.04282/) (2022)

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

- [WorldScribe: Towards Context-Aware Live Visual Descriptions](https://mlsanity.com/paper/2408.06627/) (2024)
- [Beyond Visual Perception: Insights from Smartphone Interaction of Visually Impaired Users with Large Multimodal Models](https://mlsanity.com/paper/2502.16098/) (2025)
- [Understanding How Blind Users Handle Object Recognition Errors: Strategies and Challenges](https://mlsanity.com/paper/2408.03303/) (2024)
- ["It's trained by non-disabled people": Evaluating How Image Quality Affects Product Captioning with Vision-Language Models](https://mlsanity.com/paper/2511.08917/) (2025)
- [Probing the Gaps in ChatGPT Live Video Chat for Real-World Assistance for People who are Blind or Visually Impaired](https://mlsanity.com/paper/2508.03651/) (2025)
- [AccessShare: Co-designing Data Access and Sharing with Blind People](https://mlsanity.com/paper/2407.19351/) (2024)

## Abstract (arXiv)

Iteration of training and evaluating a machine learning model is an important process to improve its performance. However, while teachable interfaces enable blind users to train and test an object recognizer with photos taken in their distinctive environment, accessibility of training iteration and evaluation steps has received little attention. Iteration assumes visual inspection of the training photos, which is inaccessible for blind users. We explore this challenge through MyCam, a mobile app that incorporates automatically estimated descriptors for non-visual access to the photos in the users' training sets. We explore how blind participants (N=12) interact with MyCam and the descriptors through an evaluation study in their homes. We demonstrate that the real-time photo-level descriptors enabled blind users to reduce photos with cropped objects, and that participants could add more variations by iterating through and accessing the quality of their training sets. Also, Participants found the app simple to use indicating that they could effectively train it and that the descriptors were useful. However, subjective responses were not reflected in the performance of their models, partially due to little variation in training and cluttered backgrounds.