We propose a robust estimator that estimates an accurate inlier scale. The proposed method first carries out an analysis to figure out the residual distribution model using the obvious case-dependent constraint, the residual function. Then the proposed inlier scale estimator performs a global search for the scale producing the residual distribution that best fits the residual distribution model. Knowledge about the residual distribution model provides a major advantage that allows us to estimate the inlier scale correctly, thereby improving the estimation robustness.

• Trung Thanh Ngo, Hajime Nagahara, Ryusuke Sagawa, Yasuhiro Mukaigawa, Masahiko Yachida, Yasushi Yagi,
  Adaptive-Scale Robust Estimator using Distribution Model Fitting,
  In Proc. 9th Asian Conference on Computer Vision, Xi'an China, Sept., 2009
• Trung Thanh Ngo, Hajime Nagahara, Ryusuke Sagawa, Yasuhiro Mukaigawa, Masahiko Yachida, Yasushi Yagi,
  Highly Robust Estimator Using a Case-dependent Residual Distribution Model,
  IPSJ Transactions on Computer Vision and Applications, vol.1, pp.260–276, Nov., 2009

A simple version of the above method when the residual distribution is assumed Gaussian.

• Trung Thanh Ngo, Hajime Nagahara , Ryusuke Sagawa, Yasuhiro Mukaigawa, Masahiko Yachida, Yasushi Yagi,
  An Adaptive-Scale Robust Estimator for Motion Estimation,
  In Proc. 2009 IEEE International Conference on Robotics and Automation, Kobe, Japan, May 12-17, 2009.