Boundary Detection Benchmarking: Beyond F-Measures

By Xiaodi Hou, Alan L. Yuille, Christof Koch. 2013

52 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Evaluation & Benchmarks

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

In Evaluation & Benchmarks

Citations in the corpus

Holistically-Nested Edge Detection (2015, background): “…and/or learned features [26, 5], (2) multi-scale response fusion [38, 30, 28], (3) engagement of different levels of visual perception [16, 25, 37, 15] such as mid-level Gestalt law information [7], (4) structural information [6] and context [36], (5) holistic image prediction [23], (6) 3D…”
N4-Fields: Neural Network Nearest Neighbor Fields for Image Transforms (2014, methodology): “Following [27] in order to account for the inherent problems of the dataset we also test our approach against the so-called “consensus” subset of the ground-truth labels.”
Crisp Boundary Detection Using Pointwise Mutual Information (2014, background): “[1] pointed out that many of the “ground truth” contours in BSDS are perceptually quite weak, and may not be indicative of real object boundaries.”
PASCAL Boundaries: A Class-Agnostic Semantic Boundary Dataset (2015, background): “Hou et al. [13] discuss the ambiguities in these datasets in more detail.”
Visual Boundary Prediction: A Deep Neural Prediction Network and Quality Dissection (2014, background): “Considering separately the performace for strong boundaries we find in agreement with Hou et al. [2013] that all methods perform significantly worse at predicting strong boundaries only (compared to predicting the full set of annotations); this is explained by the fact that many strong (but not cons”
Weakly Supervised Object Boundaries (2015, background): “These are so called “consensus” annotations [18], and correspond to sparse ∼15% of all true positives.”