Deformable Part-Based Model Transfer for Object Detection
スポンサーリンク
概要
- 論文の詳細を見る
The transfer of prior knowledge from source domains can improve the performance of learning when the training data in a target domain are insufficient. In this paper we propose a new strategy to transfer deformable part models (DPMs) for object detection, using offline-trained auxiliary DPMs of similar categories as source models to improve the performance of the target object detector. A DPM presents an object by using a root filter and several part filters. We use these filters of the auxiliary DPMs as prior knowledge and adapt the filters to the target object. With a latent transfer learning method, appropriate local features are extracted for the transfer of part filters. Our experiments demonstrate that this strategy can lead to a detector superior to some state-of-the-art methods.
- The Institute of Electronics, Information and Communication Engineersの論文
The Institute of Electronics, Information and Communication Engineers | 論文
- Compensation Effect of Quasi-Inverse Filter (QIF) on Frequency Characteristic Distortion in Wideband Systems
- Subblock Processing for Frequency-Domain Turbo Equalization under Fast Fading Environments
- Measurement-Based Performance Evaluation of Coded MIMO-OFDM Spatial Multiplexing with MMSE Spatial Filtering in an Indoor Line-of-Sight Environment
- Design of a Multiple-Input SC DC-DC Converter Realizing Long Battery Runtime
- The Influence of a Low-Level Color or Figure Adaptation on a High-Level Face Perception