English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Robust estimation is a crucial and still challenging task, which involvesestimating model parameters in noisy environments. Although conventionalsampling consensus-based algorithms sample several times to achieve robustness,these algorithms cannot use data features and historical informationeffectively. In this paper, we propose RLSAC, a novel Reinforcement Learningenhanced SAmple Consensus framework for end-to-end robust estimation. RLSACemploys a graph neural network to utilize both data and memory features toguide exploring directions for sampling the next minimum set. The feedback ofdownstream tasks serves as the reward for unsupervised training. Therefore,RLSAC can avoid differentiating to learn the features and the feedback ofdownstream tasks for end-to-end robust estimation. In addition, RLSACintegrates a state transition module that encodes both data and memoryfeatures. Our experimental results demonstrate that RLSAC can learn fromfeatures to gradually explore a better hypothesis. Through analysis, it isapparent that RLSAC can be easily transferred to other sampling consensus-basedrobust estimation tasks. To the best of our knowledge, RLSAC is also the firstmethod that uses reinforcement learning to sample consensus for end-to-endrobust estimation. We release our codes at https://github.com/IRMVLab/RLSAC.

RLSAC: Reinforcement Learning enhanced Sample Consensus for End-to-End Robust Estimation