# GeminiFusion: Efficient Pixel-wise Multimodal Fusion for Vision Transformer

arXiv: https://arxiv.org/abs/2406.01210 · HTML: https://mlsanity.com/paper/2406.01210/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Builds on 55 of its 75 references that are in the map.
- First posted on arXiv 3 Jun 2024.

## Builds on (most cited first, 30 of 55)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://mlsanity.com/paper/2105.15203/) (2021)
- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks](https://mlsanity.com/paper/1908.02265/) (2019)
- [BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers](https://mlsanity.com/paper/2203.17270/) (2022)
- [Multimodal Machine Learning: A Survey and Taxonomy](https://mlsanity.com/paper/1705.09406/) (2017)
- [VL-BERT: Pre-training of Generic Visual-Linguistic Representations](https://mlsanity.com/paper/1908.08530/) (2019)
- [VideoBERT: A Joint Model for Video and Language Representation Learning](https://mlsanity.com/paper/1904.01766/) (2019)
- [Taskonomy: Disentangling Task Transfer Learning](https://mlsanity.com/paper/1804.08328/) (2018)
- [ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks](https://mlsanity.com/paper/1910.03151/) (2019)
- [Multi-Modal Fusion Transformer for End-to-End Autonomous Driving](https://mlsanity.com/paper/2104.09224/) (2021)
- [Attention Bottlenecks for Multimodal Fusion](https://mlsanity.com/paper/2107.00135/) (2021)
- [What Makes Training Multi-Modal Classification Networks Hard?](https://mlsanity.com/paper/1905.12681/) (2019)
- [Audio-Visual Scene Analysis with Self-Supervised Multisensory Features](https://mlsanity.com/paper/1804.03641/) (2018)
- [CMT: Convolutional Neural Networks Meet Vision Transformers](https://mlsanity.com/paper/2107.06263/) (2021)
- [Cross-Modal Self-Attention Network for Referring Image Segmentation](https://mlsanity.com/paper/1904.04745/) (2019)
- [CMX: Cross-Modal Fusion for RGB-X Semantic Segmentation with Transformers](https://mlsanity.com/paper/2203.04838/) (2022)
- [EPIC-Fusion: Audio-Visual Temporal Binding for Egocentric Action Recognition](https://mlsanity.com/paper/1908.08498/) (2019)
- [Omnivore: A Single Model for Many Visual Modalities](https://mlsanity.com/paper/2201.08377/) (2022)
- [Visual Saliency Transformer](https://mlsanity.com/paper/2104.12099/) (2021)
- [Modulating early visual processing by language](https://mlsanity.com/paper/1707.00683/) (2017)
- [MVX-Net: Multimodal VoxelNet for 3D Object Detection](https://mlsanity.com/paper/1904.01649/) (2019)
- [MUTAN: Multimodal Tucker Fusion for Visual Question Answering](https://mlsanity.com/paper/1705.06676/) (2017)
- [Multimodal Token Fusion for Vision Transformers](https://mlsanity.com/paper/2204.08721/) (2022)
- [Learning Rich Features from RGB-D Images for Object Detection and Segmentation](https://mlsanity.com/paper/1407.5736/) (2014)

## Built on by (most cited first, 30 of 44)

- [Multimodal Alignment and Fusion: A Survey](https://mlsanity.com/paper/2411.17040/) (2024)
- [CAFuser: Condition-Aware Multimodal Fusion for Robust Semantic Perception of Driving Scenes](https://mlsanity.com/paper/2410.10791/) (2024)
- [DFormerv2: Geometry Self-Attention for RGBD Semantic Segmentation](https://mlsanity.com/paper/2504.04701/) (2025)
- [Benchmarking Multi-modal Semantic Segmentation under Sensor Failures: Missing and Noisy Modality Robustness](https://mlsanity.com/paper/2503.18445/) (2025)
- [Unveiling the Potential of Segment Anything Model 2 for RGB-Thermal Semantic Segmentation with Language Guidance](https://mlsanity.com/paper/2503.02581/) (2025)
- [Hyperdimensional Uncertainty Quantification for Multimodal Uncertainty Fusion in Autonomous Vehicles Perception](https://mlsanity.com/paper/2503.20011/) (2025)
- [Multimodal-Aware Fusion Network for Referring Remote Sensing Image Segmentation](https://mlsanity.com/paper/2503.11183/) (2025)
- [DGFusion: Depth-Guided Sensor Fusion for Robust Semantic Perception](https://mlsanity.com/paper/2509.09828/) (2025)
- [Cross-Modal Interactive Perception Network with Mamba for Lung Tumor Segmentation in PET-CT Images](https://mlsanity.com/paper/2503.17261/) (2025)
- [MANGO: Multimodal Attention-based Normalizing Flow Approach to Fusion Learning](https://mlsanity.com/paper/2508.10133/) (2025)
- [ADEM-VL: Adaptive and Embedded Fusion for Efficient Vision-Language Tuning](https://mlsanity.com/paper/2410.17779/) (2024)
- [DGIQA: Depth-guided Feature Attention and Refinement for Generalizable Image Quality Assessment](https://mlsanity.com/paper/2505.24002/) (2025)
- [DiffPixelFormer: Differential Pixel-Aware Transformer for RGB-D Indoor Scene Segmentation](https://mlsanity.com/paper/2511.13047/) (2025)
- [Diffusion-based RGB-D Semantic Segmentation with Deformable Attention Transformer](https://mlsanity.com/paper/2409.15117/) (2024)
- [Semantic Segmentation and Scene Reconstruction of RGB-D Image Frames: An End-to-End Modular Pipeline for Robotic Applications](https://mlsanity.com/paper/2410.17988/) (2024)
- [IV-tuning: Parameter-Efficient Transfer Learning for Infrared-Visible Tasks](https://mlsanity.com/paper/2412.16654/) (2024)
- [Learning Flow-Guided Registration for RGB-Event Semantic Segmentation](https://mlsanity.com/paper/2505.01548/) (2025)
- [Dual Atrous Separable Convolution for Improving Agricultural Semantic Segmentation](https://mlsanity.com/paper/2506.22570/) (2025)
- [DMTrack: Spatio-Temporal Multimodal Tracking via Dual-Adapter](https://mlsanity.com/paper/2508.01592/) (2025)
- [Multi-modal Uncertainty Robust Tree Cover Segmentation For High-Resolution Remote Sensing Images](https://mlsanity.com/paper/2509.04870/) (2025)
- [Multimodal SAM-adapter for Semantic Segmentation](https://mlsanity.com/paper/2509.10408/) (2025)
- [Re-coding for Uncertainties: Edge-awareness Semantic Concordance for Resilient Event-RGB Segmentation](https://mlsanity.com/paper/2511.08269/) (2025)
- [GeomPrompt: Geometric Prompt Learning for RGB-D Semantic Segmentation Under Missing and Degraded Depth](https://mlsanity.com/paper/2604.11585/) (2026)
- [RSGMamba: Reliability-Aware Self-Gated State Space Model for Multimodal Semantic Segmentation](https://mlsanity.com/paper/2604.12319/) (2026)
- [Structure-Semantic Decoupled Modulation of Global Geospatial Embeddings for High-Resolution Remote Sensing Mapping](https://mlsanity.com/paper/2604.19591/) (2026)
- [Vanishing Depth: Training Generalized Depth Adapters with Sinusoidal Depth Preprocessing for Pretrained RGB Encoders](https://mlsanity.com/paper/2503.19947/) (2025)
- [RLBayes: a Bayesian Network Structure Learning Algorithm via Reinforcement Learning-Based Search Strategy](https://mlsanity.com/paper/2504.05167/) (2025)
- [DepthMatch: Semi-Supervised RGB-D Scene Parsing through Depth-Guided Regularization](https://mlsanity.com/paper/2505.20041/) (2025)
- [DCIRNet: Depth Completion with Iterative Refinement for Dexterous Grasping of Transparent and Reflective Objects](https://mlsanity.com/paper/2506.09491/) (2025)
- [HiddenObject: Modality-Agnostic Fusion for Multimodal Hidden Object Detection](https://mlsanity.com/paper/2508.21135/) (2025)

## Abstract (arXiv)

Cross-modal transformers have demonstrated superiority in various vision tasks by effectively integrating different modalities. This paper first critiques prior token exchange methods which replace less informative tokens with inter-modal features, and demonstrate exchange based methods underperform cross-attention mechanisms, while the computational demand of the latter inevitably restricts its use with longer sequences. To surmount the computational challenges, we propose GeminiFusion, a pixel-wise fusion approach that capitalizes on aligned cross-modal representations. GeminiFusion elegantly combines intra-modal and inter-modal attentions, dynamically integrating complementary information across modalities. We employ a layer-adaptive noise to adaptively control their interplay on a per-layer basis, thereby achieving a harmonized fusion process. Notably, GeminiFusion maintains linear complexity with respect to the number of input tokens, ensuring this multimodal framework operates with efficiency comparable to unimodal networks. Comprehensive evaluations across multimodal image-to-image translation, 3D object detection and arbitrary-modal semantic segmentation tasks, including RGB, depth, LiDAR, event data, etc. demonstrate the superior performance of our GeminiFusion against leading-edge techniques. The PyTorch code is available at https://github.com/JiaDingCN/GeminiFusion