Adaptive Context Network for Scene Parsing

By Jun Fu, Jing Liu, Yuhang Wang, Yong Li, Yongjun Bao, Jinhui Tang, Hanqing Lu. 2019

149 citations recorded by the source corpus.

Recent works attempt to improve scene parsing performance by exploring different levels of contexts, and typically train a well-designed convolutional network to exploit useful contexts across all pixels equally. However, in this paper, we find that the context demands are varying from different pixels or regions in each image. Based on this observation, we propose an Adaptive Context Network (ACNet) to capture the pixel-aware contexts by a competitive fusion of global context and local context according to different per-pixel demands. Specifically, when given a pixel, the global context demand is measured by the similarity between the global feature and its local feature, whose reverse value can also be used to measure the local context demand. We model the two demanding measurements by the proposed global context module and local context module, respectively, to generate their adaptive contextual features. Furthermore, we import multiple such modules to build several adaptive context blocks in different levels of network to obtain a coarse-to-fine result. Finally, comprehensive experimental evaluations demonstrate the effectiveness of the proposed ACNet, and new state-of-the-arts performances are achieved on all four public datasets, i.e. Cityscapes, ADE20K, PASCAL Context, and COCO Stuff.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Language-driven Semantic Segmentation (2022, neutral): “, 2020), ACNet (Fu et al., 2019), DeeplabV3 (Chen et al.”
Object-Contextual Representations for Semantic Segmentation (2019, background): “Some other studies [7,42,17] construct the encoder-decoder structures to exploit the multi-resolution features as the multi-scale context.”
Segmenter: Transformer for Semantic Segmentation (2021, neutral): “…and enable fast inference with 400 images per second for Seg-B-Mask/32, four Method Backbone Im/sec mIoU +MS OCR [60] HRNetV2-W48 83 - 45.66 ACNet [24] ResNet-101 - - 45.90 DNL [57] ResNet-101 - - 45.97 DRANet [22] ResNet-101 - - 46.18 CPNet [58] ResNet-101 - - 46.27 DeepLabv3+ [10] ResNet-101 76…”
ResNeSt: Split-Attention Networks (2020, neutral): “d Backbone pixAcc% mIoU% ork UperNet [59] ResNet101 81.01 42.66 PSPNet [69] ResNet101 81.39 43.29 EncNet [65] ResNet101 81.69 44.65 CFNet [66] ResNet101 81.57 44.89 OCNet [63] ResNet101 - 45.45 ACNet [17] ResNet101 81.96 45.90 Ours DeeplabV3 [7] ResNet50 [21] 80.39 42.1 ResNet101 [21] 81.11 44.14 Re”
Hierarchical Multi-Scale Attention for Semantic Segmentation (2020, background): “More recently, ZigZagNet [7] and ACNet [8] leverage intermediate features instead of just the features from the ﬁnal layer of the network trunk to create the multi-scale context.”
Disentangled Non-Local Neural Networks (2020, unknown): “48 42.99 - 54.0 EncNet [41] ResNet-101 44.65 55.67 52.6 DANet [13] ResNet-101 45.22 - 52.6 CFNet [42] ResNet-101 44.89 - 54.0 ANN [47] ResNet-101 45.24 - 52.8 DMNet [17] ResNet-101 45.50 - 54.4 ACNet [14] ResNet-101 45.90 55.84 54.1 NL ResNet-101 44.67 55.58 50.6 DNL (ours) ResNet-101 45.97 56.23 54”
Bi-directional Cross-Modality Feature Propagation with Separation-and-Aggregation Gate for RGB-D Semantic Segmentation (2020, unknown): “Besides exploiting various contextual information from the visual cues [25,11,14,12,5,43], depth data have recently been utilized as supplementary information to RGB data to achieve improved segmentation accuracy [26,33,44,4,15,23,6,19].”
OCNet: Object Context for Semantic Segmentation (2021, methodology): “We ﬁrst apply the self-attention scheme to predict the dense relation matrix and verify its capability to approximate the object context, and there also exist some concurrent studies (Fu et al. 2019a; Zhang et al. 2019b) that apply the self-attention scheme for semantic segmentation.”
Dynamic Multi-Scale Filters for Semantic Segmentation (2019, methodology): “Finally, we use averaged predictions of different scales as final prediction [30, 51, 38, 46, 17, 10, 20, 12, 11] and adopt mean of classwise intersection over union (mIoU) as our evaluation metric .”
Adaptive Context Selection for Polyp Segmentation (2020, background): “Inspired by [4], we believe that the global context features are helpful for the segmentation of large polyps, while the local context information is crucial for the identification of small ones.”