See the Difference: Direct Pre-Image Reconstruction and Pose Estimation by Differentiating HOG
Authors:
- Wei-Chen ChiuWCWei-Chen Chiu
- Mario FritzMFMario Fritz
- 1 December 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 468-476
Abstract
The Histogram of Oriented Gradient (HOG) descriptor has led to many advances in computer vision over the last decade and is still part of many state of the art approaches. We realize that the associated feature computation is piecewise differentiable and therefore many pipelines which build on HOG can be made differentiable. This lends to advanced introspection as well as opportunities for end-to-end optimization. We present our implementation of ΔHOG based on the auto-differentiation toolbox Chumpy [18] and show applications to pre-image visualization and pose estimation which extends the existing differentiable renderer OpenDR [19] pipeline. Both applications improve on the respective state-of-the-art HOG approaches.Keywords
All Related Versions
This publication has 24 references indexed in Scilit:
- Picture: A probabilistic programming language for scene perception2015
- Image Reconstruction from Bag-of-Visual-Words2014
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014
- Beyond PASCAL: A benchmark for 3D object detection in the wild2014
- HOGgles: Visualizing Object Detection Features2013
- Parsing IKEA Objects: Fine Pose Estimation2013
- Supervised Descent Method and Its Applications to Face Alignment2013
- Reconstructing an image from its local descriptors2011
- Local isomorphism to solve the pre-image problem in kernel methods2011
- Histograms of Oriented Gradients for Human Detection2005