Learning Deep Multimodal Feature Representation with Asymmetric Multi-layer Fusion
This work addresses the challenge of efficiently fusing multimodal data for computer vision tasks, offering a more compact and effective solution compared to existing methods, though it is incremental in nature.
The authors tackled the problem of multimodal feature fusion by proposing a compact framework that uses a shared network with modality-specific batch normalization and bidirectional multi-layer fusion with asymmetric operations, achieving superior performance to state-of-the-art methods on semantic segmentation and image translation tasks across three datasets.
We propose a compact and effective framework to fuse multimodal features at multiple layers in a single network. The framework consists of two innovative fusion schemes. Firstly, unlike existing multimodal methods that necessitate individual encoders for different modalities, we verify that multimodal features can be learnt within a shared single network by merely maintaining modality-specific batch normalization layers in the encoder, which also enables implicit fusion via joint feature representation learning. Secondly, we propose a bidirectional multi-layer fusion scheme, where multimodal features can be exploited progressively. To take advantage of such scheme, we introduce two asymmetric fusion operations including channel shuffle and pixel shift, which learn different fused features with respect to different fusion directions. These two operations are parameter-free and strengthen the multimodal feature interactions across channels as well as enhance the spatial feature discrimination within channels. We conduct extensive experiments on semantic segmentation and image translation tasks, based on three publicly available datasets covering diverse modalities. Results indicate that our proposed framework is general, compact and is superior to state-of-the-art fusion frameworks.