# Learning high-level visual representations from a child's perspective without strong inductive biases

arXiv: https://arxiv.org/abs/2305.15372 · HTML: https://mlsanity.com/paper/2305.15372/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 60 references that are in the map.
- First posted on arXiv 24 May 2023.

## Builds on (most cited first, 22 of 22)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Aggregated Residual Transformations for Deep Neural Networks](https://mlsanity.com/paper/1611.05431/) (2016)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Ego4D: Around the World in 3,000 Hours of Egocentric Video](https://mlsanity.com/paper/2110.07058/) (2021)
- [The 2017 DAVIS Challenge on Video Object Segmentation](https://mlsanity.com/paper/1704.00675/) (2017)
- [Object-Centric Learning with Slot Attention](https://mlsanity.com/paper/2006.15055/) (2020)
- [The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scale](https://mlsanity.com/paper/1811.00982/) (2018)
- [CORe50: a New Dataset and Benchmark for Continuous Object Recognition](https://mlsanity.com/paper/1705.03550/) (2017)
- [Space-Time Correspondence as a Contrastive Random Walk](https://mlsanity.com/paper/2006.14613/) (2020)
- [Partial success in closing the gap between human and machine vision](https://mlsanity.com/paper/2106.07411/) (2021)
- [A Short Note on the Kinetics-700-2020 Human Action Dataset](https://mlsanity.com/paper/2010.10864/) (2020)
- [Self-supervised learning through the eyes of a child](https://mlsanity.com/paper/2007.16189/) (2020)
- [Mugs: A Multi-Granular Self-Supervised Learning Framework](https://mlsanity.com/paper/2203.14415/) (2022)
- [The developmental trajectory of object recognition robustness: children are like small adults but unlike big deep neural networks](https://mlsanity.com/paper/2205.10144/) (2022)
- [Controlled-rearing studies of newborn chicks and deep neural networks](https://mlsanity.com/paper/2112.06106/) (2021)

## Built on by (most cited first, 22 of 22)

- [The BabyView dataset: High-resolution egocentric videos of infants' and young children's everyday experiences](https://mlsanity.com/paper/2406.10447/) (2024)
- [Emulating Human-like Adaptive Vision for Efficient and Flexible Machine Visual Perception](https://mlsanity.com/paper/2509.15333/) (2025)
- [Better artificial intelligence does not mean better models of biology](https://mlsanity.com/paper/2504.16940/) (2025)
- [Whither symbols in the era of advanced neural networks?](https://mlsanity.com/paper/2508.05776/) (2025)
- [Self-supervised learning of video representations from a child's perspective](https://mlsanity.com/paper/2402.00300/) (2024)
- [BabyVLM: Data-Efficient Pretraining of VLMs Inspired by Infant Learning](https://mlsanity.com/paper/2504.09426/) (2025)
- [Adopting a human developmental visual diet yields robust, shape-based AI vision](https://mlsanity.com/paper/2507.03168/) (2025)
- [On the robustness of modeling grounded word learning through a child's egocentric input](https://mlsanity.com/paper/2507.14749/) (2025)
- [Overcoming classic challenges for artificial neural networks by providing incentives and practice](https://mlsanity.com/paper/2410.10596/) (2024)
- [Learning to See Through a Baby's Eyes: Early Visual Diets Enable Robust Visual Intelligence in Humans and Machines](https://mlsanity.com/paper/2511.14440/) (2025)
- [Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes](https://mlsanity.com/paper/2608.05000/) (2026)
- [HVM-1: Large-scale video models pretrained with nearly 5000 hours of human-like video data](https://mlsanity.com/paper/2407.18067/) (2024)
- [Discovering Hidden Visual Concepts Beyond Linguistic Input in Infant Learning](https://mlsanity.com/paper/2501.05205/) (2025)
- [Human Gaze Boosts Object-Centered Representation Learning](https://mlsanity.com/paper/2501.02966/) (2025)
- [Toddlers' Active Gaze Behavior Supports Self-Supervised Object Learning](https://mlsanity.com/paper/2411.01969/) (2024)
- [Can Large Language Models (LLMs) Describe Pictures Like Children? A Comparative Corpus Study](https://mlsanity.com/paper/2508.13769/) (2025)
- [Simulated Cortical Magnification Supports Self-Supervised Object Learning](https://mlsanity.com/paper/2509.15751/) (2025)
- [Do Blind Spots Matter for Word-Referent Mapping? A Computational Study with Infant Egocentric Video](https://mlsanity.com/paper/2511.11725/) (2025)
- [Opinion: Learning Intuitive Physics May Require More than Visual Data](https://mlsanity.com/paper/2512.06232/) (2025)
- [Temporal Slowness in Central Vision Drives Semantic Object Learning](https://mlsanity.com/paper/2602.04462/) (2026)
- [Characterizing the visual representation of objects from the child's view](https://mlsanity.com/paper/2605.14990/) (2026)
- [Objects Before Words: Object-First Inductive Biases for Grounding Language in Child-View Video](https://mlsanity.com/paper/2606.12985/) (2026)

## Abstract (arXiv)

Young children develop sophisticated internal models of the world based on their visual experience. Can such models be learned from a child's visual experience without strong inductive biases? To investigate this, we train state-of-the-art neural networks on a realistic proxy of a child's visual experience without any explicit supervision or domain-specific inductive biases. Specifically, we train both embedding models and generative models on 200 hours of headcam video from a single child collected over two years and comprehensively evaluate their performance in downstream tasks using various reference models as yardsticks. On average, the best embedding models perform at a respectable 70% of a high-performance ImageNet-trained model, despite substantial differences in training data. They also learn broad semantic categories and object localization capabilities without explicit supervision, but they are less object-centric than models trained on all of ImageNet. Generative models trained with the same data successfully extrapolate simple properties of partially masked objects, like their rough outline, texture, color, or orientation, but struggle with finer object details. We replicate our experiments with two other children and find remarkably consistent results. Broadly useful high-level visual representations are thus robustly learnable from a representative sample of a child's visual experience without strong inductive biases.