Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection

Akyön, Fatih
Altınuç, Sinan Onur
Temizel, Alptekin
IEEE International Conference on Image Processing (ICIP)


Depth assisted object segmentation in multi-view video
Cigla, Cevahir; Alatan, Abdullah Aydın (2008-01-01)
In this work, a novel and unified approach for multi-view video (MVV) object segmentation is presented. In the first stage, a region-based graph-theoretic color segmentation algorithm is proposed, in which the popular Normalized Cuts segmentation method is improved with some modifications on its graph structure. Segmentation is obtained by recursive bi-partitioning of a weighted graph of an initial over-segmentation mask. The available segmentation mask is also utilized during dense depth map estimation ste...
Scheduling for flexible layout.
Süer, Ahmet Gürsel; Department of Industrial Engineering (1985)
Waterfall region analysis for iterative decoding
Yılmaz, Ali Özgür (2004-12-01)
Finite length analysis of iterative decoders can be done by using probabilistic models based on EXIT charts. The validity of these models will be investigated by checking the performance of iterative decoding under various scenarios.
Querying Class-Relationship Logic in a Metalogic Framework
Nilsson, Jorgen Fischer (2011-10-28)
We introduce a class relationship logic for stating various forms of logical relationships between classes. This logic is intended for ontologies and knowledge bases and combinations thereof. Reasoning and querying is conducted in the DATALOC logical language, which serves as an embracing decidable and tractable metalogic.
Slice coherence in a query-based architecture for 3D heterogeneous printing
Yaman, Ulaş; Sacks, Elisha; Hoffmann, Christoph (2016-06-01)
We report on 3D printing of artifacts with a structured, inhomogeneous interior. The interior is decomposed into cells defined by a 3D Voronoi diagram and their sites. When printing such objects, most slices the printer deposits are topologically the same and change only locally in the interior. The slicing algorithm capitalizes on this coherence and minimizes print head moves that do not deposit material. This approach has been implemented on a client/server architecture that computes the slices on the geo...
Citation Formats
F. Akyön, S. O. Altınuç, and A. Temizel, “Slicing Aided Hyper Inference and Fine-tuning for Small Object Detection,” presented at the IEEE International Conference on Image Processing (ICIP), Bordeaux, Fransa, 2022, Accessed: 00, 2022. [Online]. Available: