Learning Dynamic Hierarchical Models for Anytime Scene Labeling
نویسندگان
چکیده
With increasing demand for efficient image and video analysis, test-time cost of scene parsing becomes critical for many large-scale or time-sensitive vision applications. We propose a dynamic hierarchical model for anytime scene labeling that allows us to achieve flexible trade-offs between efficiency and accuracy in pixel-level prediction. In particular, our approach incorporates the cost of feature computation and model inference, and optimizes the model performance for any given test-time budget by learning a sequence of image-adaptive hierarchical models. We formulate this anytime representation learning as a Markov Decision Process with a discrete-continuous state-action space. A highquality policy of feature and model selection is learned based on an approximate policy iteration method with action proposal mechanism. We demonstrate the advantages of our dynamic non-myopic anytime scene parsing on three semantic segmentation datasets, which achieves 90% of the state-of-the-art performances by using 15% of their overall costs.
منابع مشابه
Stacked Hierarchical Labeling
In this work we propose a hierarchical approach for labeling semantic objects and regions in scenes. Our approach is reminiscent of early vision literature in that we use a decomposition of the image in order to encode relational and spatial information. In contrast to much existing work on structured prediction for scene understanding, we bypass a global probabilistic model and instead directl...
متن کاملSemantic Labeling of SAR Images with Hierarchical Markov Aspect Models
Scene segmentation and semantic labeling of Synthetic Aperture Radar (SAR) images is one of the key problems in interpreting SAR data. In this paper, a new approach for semantic labeling of SAR imagery is proposed based on hierarchical Markov aspect model (HMAM) with weak supervision. The motivation for this work is to incorporate the multiscale spatial relation between adjacent image patches i...
متن کاملSAR Image Labeling with Hierarchical Markov Aspect Models
Scene segmentation and semantic labeling are important problems in SAR image interpretation. This paper proposes an efficient SAR imagery labeling method based on aspect model which can be learnt from keywords-labeled training data directly. Furthermore, a novel hierarchical Markov aspect model (HMAM) is presented by building aspect model on quadtree. HMAM outperform both aspect model and hiera...
متن کاملSpeedMachines: Anytime Structured Prediction
Structured prediction plays a central role in machine learning applications from computational biology to computer vision. These models require significantly more computation than unstructured models, and, in many applications, algorithms may need to make predictions within a computational budget or in an anytime fashion. In this work we propose an anytime technique for learning structured pred...
متن کاملHybrid Hierarchical Learning from Dynamic Scenes
The work proposes a hierarchical architecture for learning from dynamic scenes at various levels of knowledge abstraction. The raw visual information is processed at different stages to generate hybrid symbolic/sub-symbolic descriptions of the scene, agents and events. The background is incrementally learned at the lowest layer, which is used further in the mid-level for multi-agent tracking wi...
متن کامل