Laplacian Reconstruction and Refinement for Semantic Segmentation

By Golnaz Ghiasi, Charless C. Fowlkes. 2016

49 citations recorded by the source corpus.

CNN architectures have terrific recognition performance but rely on spatial pooling which makes it difficult to adapt them to tasks that require dense pixel-accurate labeling. This paper makes two contributions: (1) We demonstrate that while the apparent spatial resolution of convolutional feature maps is low, the high-dimensional feature representation contains significant sub-pixel localization information. (2) We describe a multi-resolution reconstruction architecture, akin to a Laplacian pyramid, that uses skip connections from higher resolution feature maps to successively refine segment boundaries reconstructed from lower resolution maps. This approach yields state-of-the-art semantic segmentation results on PASCAL without resorting to more complex CRF or detection driven architectures.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Rethinking Atrous Convolution for Semantic Image Segmentation (2017, methodology): “Adelaide VeryDeep FCN VOC [76] 79.1 LRR 4x ResNet-CRF [21] 79.3 DeepLabv2-CRF [10] 79.7 CentraleSupelec Deep G-CRF [7] 80.2 HikSeg COCO [71] 81.4 Deep Layer Cascade (LC) [45] 82.7 TuSimple [75] 83.1 Large Kernel Matters [60] 83.6 Multipath-ReﬁneNet [47] 84.2 ResNet-38 MS COCO [77] 84.9 PSPNet [84]…”
Path Aggregation Network for Instance Segmentation (2018, methodology): “SharpMask [49], Peng et al. [47] and LRR [14] fused feature maps for segmentation with finer details.”
UberNet: Training a Universal Convolutional Neural Network for Low-, Mid-, and High-Level Vision Using Diverse Datasets and Limited Memory (2016, background): “U-net type architectures [29, 75, 76, 87], Dense CRF post-processing [14, 51, 113] or bilateral filtertype smoothing [12, 39], as well as more general structured prediction with CNN-based pairwise terms [11,61] - but we leave this for future work.”
Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes (2016, methodology): “Published Unpublished LRR[20] Adelaide[34] Dilation[55] ENet[44] SegNet[2] DeepLab[43] FRRNA/B Dilation[55] LRR[20] FRRNB Figure 6.”
AdapNet: Adaptive semantic segmentation in adverse environmental conditions (2017, background): “Filters learned by DCNNs are often not well suited for this multiscale appearance.”
SGN: Sequential Grouping Networks for Instance Segmentation (2017, methodology): “Note that the LRR model we use is pre-trained on MS COCO [25], which is also used by DIN [1] for pre-training.”
Knowledge Distillation for Semantic Segmentation Using Channel and Spatial Correlations and Adaptive Cross Entropy (2020, methodology): “To deal with this, networks that combine both low- and high-level features have been proposed [8,9].”
A survey on instance segmentation: state of the art (2020, background): “The authors of [79] discuss that although the practice of using features from different layers in image recognition is well established [89,118,119,47,120,121,48,53,82,122,123].”