# MMG-Ego4D: Multi-Modal Generalization in Egocentric Action Recognition

arXiv: https://arxiv.org/abs/2305.07214 · HTML: https://mlsanity.com/paper/2305.07214/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- Builds on 58 of its 78 references that are in the map.
- First posted on arXiv 12 May 2023.

## Builds on (most cited first, 30 of 58)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [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)
- [Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks](https://mlsanity.com/paper/1703.03400/) (2017)
- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [Matching Networks for One Shot Learning](https://mlsanity.com/paper/1606.04080/) (2016)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [Ego4D: Around the World in 3,000 Hours of Egocentric Video](https://mlsanity.com/paper/2110.07058/) (2021)
- [Calibrate Before Use: Improving Few-Shot Performance of Language Models](https://mlsanity.com/paper/2102.09690/) (2021)
- [ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision](https://mlsanity.com/paper/2102.03334/) (2021)
- [A Closer Look at Spatiotemporal Convolutions for Action Recognition](https://mlsanity.com/paper/1711.11248/) (2017)
- [On First-Order Meta-Learning Algorithms](https://mlsanity.com/paper/1803.02999/) (2018)
- [Adversarial Machine Learning at Scale](https://mlsanity.com/paper/1611.01236/) (2016)
- [Multiscale Vision Transformers](https://mlsanity.com/paper/2104.11227/) (2021)
- [Scaling Egocentric Vision: The EPIC-KITCHENS Dataset](https://mlsanity.com/paper/1804.02748/) (2018)
- [Multimodal Few-Shot Learning with Frozen Language Models](https://mlsanity.com/paper/2106.13884/) (2021)
- [AST: Audio Spectrogram Transformer](https://mlsanity.com/paper/2104.01778/) (2021)
- [X3D: Expanding Architectures for Efficient Video Recognition](https://mlsanity.com/paper/2004.04730/) (2020)
- [Rescaling Egocentric Vision](https://mlsanity.com/paper/2006.13256/) (2020)
- [AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions](https://mlsanity.com/paper/1705.08421/) (2017)
- [VATT: Transformers for Multimodal Self-Supervised Learning from Raw Video, Audio and Text](https://mlsanity.com/paper/2104.11178/) (2021)
- [Convolutional Two-Stream Network Fusion for Video Action Recognition](https://mlsanity.com/paper/1604.06573/) (2016)
- [Attention Bottlenecks for Multimodal Fusion](https://mlsanity.com/paper/2107.00135/) (2021)
- [Meta-Transfer Learning for Few-Shot Learning](https://mlsanity.com/paper/1812.02391/) (2018)
- [Efficient Low-rank Multimodal Fusion with Modality-Specific Factors](https://mlsanity.com/paper/1806.00064/) (2018)
- [SMIL: Multimodal Learning with Severely Missing Modality](https://mlsanity.com/paper/2103.05677/) (2021)

## Built on by (most cited first, 26 of 26)

- [RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Robotics](https://mlsanity.com/paper/2411.16537/) (2024)
- [Deep Multimodal Learning with Missing Modality: A Survey](https://mlsanity.com/paper/2409.07825/) (2024)
- [EgoVid-5M: A Large-Scale Video-Action Dataset for Egocentric Video Generation](https://mlsanity.com/paper/2411.08380/) (2024)
- [Challenges and Trends in Egocentric Vision: A Survey](https://mlsanity.com/paper/2503.15275/) (2025)
- [Masked Video and Body-worn IMU Autoencoder for Egocentric Action Recognition](https://mlsanity.com/paper/2407.06628/) (2024)
- [Multimodal Cross-Domain Few-Shot Learning for Egocentric Action Recognition](https://mlsanity.com/paper/2405.19917/) (2024)
- [EMHI: A Multimodal Egocentric Human Motion Dataset with HMD and Body-Worn IMUs](https://mlsanity.com/paper/2408.17168/) (2024)
- [COMODO: Cross-Modal Video-to-IMU Distillation for Efficient Egocentric Human Activity Recognition](https://mlsanity.com/paper/2503.07259/) (2025)
- [EMAG: Ego-motion Aware and Generalizable 2D Hand Forecasting from Egocentric Videos](https://mlsanity.com/paper/2405.20030/) (2024)
- [Exploring Missing Modality in Multimodal Egocentric Datasets](https://mlsanity.com/paper/2401.11470/) (2024)
- [Cognition Transferring and Decoupling for Text-supervised Egocentric Semantic Segmentation](https://mlsanity.com/paper/2410.01341/) (2024)
- [ANNEXE: Unified Analyzing, Answering, and Pixel Grounding for Egocentric Interaction](https://mlsanity.com/paper/2504.01472/) (2025)
- [Reconstruct before Query: Continual Missing Modality Learning with Decomposed Prompt Collaboration](https://mlsanity.com/paper/2403.11373/) (2024)
- [MANGO: Multimodal Attention-based Normalizing Flow Approach to Fusion Learning](https://mlsanity.com/paper/2508.10133/) (2025)
- [C3T: Cross-modal Transfer Through Time for Sensor-based Human Activity Recognition](https://mlsanity.com/paper/2407.16803/) (2024)
- [Integrating Audio Narrations to Strengthen Domain Generalization in Multimodal First-Person Action Recognition](https://mlsanity.com/paper/2409.09611/) (2024)
- [Egocentric Action-aware Inertial Localization in Point Clouds with Vision-Language Guidance](https://mlsanity.com/paper/2505.14346/) (2025)
- [Multimodal Knowledge Distillation for Egocentric Action Recognition Robust to Missing Modalities](https://mlsanity.com/paper/2504.08578/) (2025)
- [Improving Out-of-distribution Human Activity Recognition via IMU-Video Cross-modal Representation Learning](https://mlsanity.com/paper/2507.13482/) (2025)
- [Towards Achieving Perfect Multimodal Alignment](https://mlsanity.com/paper/2503.15352/) (2025)
- [EgoEMS: A High-Fidelity Multimodal Egocentric Dataset for Cognitive Assistance in Emergency Medical Services](https://mlsanity.com/paper/2511.09894/) (2025)
- [MobiDiary: Autoregressive Action Captioning with Wearable Devices and Wireless Signals](https://mlsanity.com/paper/2601.08204/) (2026)
- [SAVA-X: Ego-to-Exo Imitation Error Detection via Scene-Adaptive View Alignment and Bidirectional Cross View Fusion](https://mlsanity.com/paper/2603.12764/) (2026)
- [MAND: Modality-Aware Novelty Detection for Open-World Egocentric Activity Recognition](https://mlsanity.com/paper/2603.16970/) (2026)
- [UNIEGO: Proxies as Mediators for Unified Egocentric Video Representation Learning](https://mlsanity.com/paper/2606.20559/) (2026)
- [EgoInertia-MI: A Multimodal Egocentric Vision and IMU Benchmark for Motor Impairment Assessment](https://mlsanity.com/paper/2607.03934/) (2026)

## Abstract (arXiv)

In this paper, we study a novel problem in egocentric action recognition, which we term as "Multimodal Generalization" (MMG). MMG aims to study how systems can generalize when data from certain modalities is limited or even completely missing. We thoroughly investigate MMG in the context of standard supervised action recognition and the more challenging few-shot setting for learning new action categories. MMG consists of two novel scenarios, designed to support security, and efficiency considerations in real-world applications: (1) missing modality generalization where some modalities that were present during the train time are missing during the inference time, and (2) cross-modal zero-shot generalization, where the modalities present during the inference time and the training time are disjoint. To enable this investigation, we construct a new dataset MMG-Ego4D containing data points with video, audio, and inertial motion sensor (IMU) modalities. Our dataset is derived from Ego4D dataset, but processed and thoroughly re-annotated by human experts to facilitate research in the MMG problem. We evaluate a diverse array of models on MMG-Ego4D and propose new methods with improved generalization ability. In particular, we introduce a new fusion module with modality dropout training, contrastive-based alignment training, and a novel cross-modal prototypical loss for better few-shot performance. We hope this study will serve as a benchmark and guide future research in multimodal generalization problems. The benchmark and code will be available at https://github.com/facebookresearch/MMG_Ego4D.