Comparative Evaluation of Binary Features

By Jared Heinly, Enrique Dunn, Jan-Michael Frahm. European Conference on Computer Vision, 2012

370 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/d7f016f6ceb87092bc5cff84fafd94bfe3fa4adf

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

MatchNet: Unifying feature and metric learning for patch-based matching (2015, background): “Early efforts focused on identifying better affine region detectors [16], engineering more robust local descriptors [7, 15], and exploring im-”
A review of multimodal image matching: Methods and applications (2021, methodology): “43 by local descriptors due to their nonclustered properties [472].”
LoFTR: Detector-Free Local Feature Matching with Transformers (2021, methodology): “, matching score or recall [13, 30]) for detector-free image matching methods.”
Comparative Evaluation of Hand-Crafted and Learned Local Features (2017, methodology): “[16] for more details and an in-depth motivation of these metrics.”
Fast Explicit Diffusion for Accelerated Features in Nonlinear Scale Spaces (2013, methodology): “Table 1 shows average combined detection and description performance results considering the Matching Score (MS) and Re-call (RC) as described in [8].”
Image Matching from Handcrafted to Deep Features: A Survey (2020, background): “At the same time, Heinly et al. (2012) compared the traditional float and binary feature operators in 2012 and evaluated theirmatching performancewith the inter-combination of existing detectors and descriptors on the public and their own datasets.”
Automatic Classification of Defective Photovoltaic Module Cells in Electroluminescence Images (2018, background): “Even though binary feature descriptors are typically very fast to compute, they generally do not perform better than real-valued descriptors [26].”
Harvesting Mid-level Visual Concepts from Large-Scale Internet Images (2013, background): “classemes [11]) however are not designed to build a dictionary of mid-level representations.”