# Towards Efficient Scene Understanding via Squeeze Reasoning

arXiv: https://arxiv.org/abs/2011.03308 · HTML: https://mlsanity.com/paper/2011.03308/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Nov 2020.

## Built on by (most cited first, 10 of 10)

- [Change Detection Methods for Remote Sensing in the Last Decade: A Comprehensive Review](https://mlsanity.com/paper/2305.05813/) (2023)
- [PointFlow: Flowing Semantics Through Points for Aerial Image Segmentation](https://mlsanity.com/paper/2103.06564/) (2021)
- [Adaptive Multi-view and Temporal Fusing Transformer for 3D Human Pose Estimation](https://mlsanity.com/paper/2110.05092/) (2021)
- [SeaFormer++: Squeeze-enhanced Axial Transformer for Mobile Visual Recognition](https://mlsanity.com/paper/2301.13156/) (2023)
- [FsaNet: Frequency Self-attention for Semantic Segmentation](https://mlsanity.com/paper/2211.15595/) (2022)
- [OV-VG: A Benchmark for Open-Vocabulary Visual Grounding](https://mlsanity.com/paper/2310.14374/) (2023)
- [Improving Scene Graph Generation with Superpixel-Based Interaction Learning](https://mlsanity.com/paper/2308.02339/) (2023)
- [Convolution-enhanced Evolving Attention Networks](https://mlsanity.com/paper/2212.08330/) (2022)
- [A Class-wise Non-salient Region Generalized Framework for Video Semantic Segmentation](https://mlsanity.com/paper/2212.14154/) (2022)
- [Local-to-Global Information Communication for Real-Time Semantic Segmentation Network Search](https://mlsanity.com/paper/2302.08481/) (2023)

## Abstract (arXiv)

Graph-based convolutional model such as non-local block has shown to be effective for strengthening the context modeling ability in convolutional neural networks (CNNs). However, its pixel-wise computational overhead is prohibitive which renders it unsuitable for high resolution imagery. In this paper, we explore the efficiency of context graph reasoning and propose a novel framework called Squeeze Reasoning. Instead of propagating information on the spatial map, we first learn to squeeze the input feature into a channel-wise global vector and perform reasoning within the single vector where the computation cost can be significantly reduced. Specifically, we build the node graph in the vector where each node represents an abstract semantic concept. The refined feature within the same semantic category results to be consistent, which is thus beneficial for downstream tasks. We show that our approach can be modularized as an end-to-end trained block and can be easily plugged into existing networks. Despite its simplicity and being lightweight, the proposed strategy allows us to establish the considerable results on different semantic segmentation datasets and shows significant improvements with respect to strong baselines on various other scene understanding tasks including object detection, instance segmentation and panoptic segmentation. Code is available at urlhttps://github.com/lxtGH/SFSegNets.