• Corpus ID: 237091320

Pixel Difference Networks for Efficient Edge Detection

  title={Pixel Difference Networks for Efficient Edge Detection},
  author={Z. Su and Wenzhe Liu and Zitong Yu and Dewen Hu and Qing Liao and Qi Tian and Matti Pietik{\"a}inen and Li Liu},
  • Z. Su, Wenzhe Liu, +5 authors Li Liu
  • Published 16 August 2021
  • Computer Science
  • ArXiv
Recently, deep Convolutional Neural Networks (CNNs) can achieve human-level performance in edge detection with the rich and abstract edge representation capacities. However, the high performance of CNN based edge detection is achieved with a large pretrained CNN backbone, which is memory and energy consuming. In addition, it is surprising that the previous wisdom from the traditional edge detectors, such as Canny, Sobel, and LBP are rarely investigated in the rapid-developing deep learning era… 
Dense Prediction with Attentive Feature Aggregation
The experimental analyses show that AFA learns to progressively refine segmentation maps and to improve boundary details, leading to new state-of-the-art results on boundary detection benchmarks on BSDS500 and NYUDv2.
MoCoPnet: Exploring Local Motion and Contrast Priors for Infrared Small Target Super-Resolution
A local spatiotemporal attention module to perform implicit frame alignment and incorporate the local spatio-temporal information to enhance the local features (especially for small targets) and a central difference residual group to incorporate the central difference convolution into the feature extraction backbone, which can achieve center-oriented gradient-aware feature extraction to further improve the target contrast.
Learning Multiple Dense Prediction Tasks from Partially Annotated Data
This paper proposes a multi-task training procedure that successfully leverages task relations to supervise its multi- task learning when data is partially annotated, and learns to map each task pair to a joint pairwise task-space which enables sharing information between them in a computationally efficient way through another network conditioned on task pairs.
SketchEdit: Mask-Free Local Image Manipulation with Partial Sketches
This work investigates a new paradigm of sketch-based image manipulation: maskfree local image manipulation, which only requires sketch inputs from users and utilizes the entire original image.


Richer Convolutional Features for Edge Detection
The proposed network fully exploits multiscale and multilevel information of objects to perform the image-to-image prediction by combining all the meaningful convolutional features in a holistic manner and achieves state-of-the-art performance on several available datasets.
Traditional Method Inspired Deep Neural Network For Edge Detection
This work simplifies the network architecture to include Feature Extractor, Enrichment, and Summarizer, which roughly correspond to gradient, low pass filter, and pixel connection in the traditional edge detection schemes, and can effectively reduce the complexity and retain the edge prediction quality.
Dense Extreme Inception Network: Towards a Robust CNN Model for Edge Detection
A Deep Learning based edge detector, which is inspired on both HED (Holistically-Nested Edge Detection) and Xception networks, which generates thin edge-maps that are plausible for human eyes; it can be used in any edge detection task without previous training or fine tuning process.
Bi-Directional Cascade Network for Perceptual Edge Detection
A Bi-Directional Cascade Network (BDCN) structure is proposed, where an individual layer is supervised by labeled edges at its specific scale, rather than directly applying the same supervision to all CNN outputs.
Learning to predict crisp boundaries
This work introduces a novel loss for boundary detection, which is very effective for classifying imbalanced data and allows CNNs to produce crisp boundaries and proposes an end-to-end network which adopts the bottom-up/top-down architecture to tackle the task.
DeepContour: A deep convolutional feature learned by positive-sharing loss for contour detection
This work shows that contour detection accuracy can be improved by instead making the use of the deep features learned from convolutional neural networks (CNNs), while rather than using the networks as a blackbox feature extractor, it customize the training strategy by partitioning contour (positive) data into subclasses and fitting each subclass by different model parameters.
FINED: Fast Inference Network for Edge Detection
A Fast Inference Network for Edge Detection (FINED) is proposed, which is a lightweight neural net dedicated to edge detection, which can achieve the state-of-the-art accuracy in edge detection while significantly reducing its complexity.
Deep Crisp Boundaries
A systematical study of ConvNet based edge detectors outputs is provided, and it is shown that they failed to accurately localize edges, which can be adversarial for tasks that require crisp edge inputs.
Object Contour Detection with a Fully Convolutional Encoder-Decoder Network
A deep learning algorithm for contour detection with a fully convolutional encoder-decoder network that generalizes well to unseen object classes from the same supercategories on MS COCO and can match state-of-the-art edge detection on BSDS500 with fine-tuning.
Learning Relaxed Deep Supervision for Better Edge Detection
  • Y. Liu, M. Lew
  • Computer Science
    2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2016
This work builds hierarchical supervisory signals with additional relaxed labels to consider the diversities in deep neural networks to use relaxed deep supervision within convolutional neural networks for edge detection and obtains superior cross-dataset generalization results on NYUD dataset.