Image Specificity ( Supplementary material
نویسندگان
چکیده
As we have shown, certain image-level objects and attributes make some images more specific than others. This means that specificity may be predictable using image features alone. To test this, a ν-SVR with an RBF kernel is trained on a randomly chosen subset of images represented by their DECAF-6 features [2] in the MEM-5S and PASCAL-50S datasets. In the ABSTRACT-50S dataset, the image features are a concatenation of object occurrence, their absolute position, depth, flip angle, object co-occurrence, and clip art category [6]. For prediction, 188 images are set aside in the MEM-5S dataset, 200 images in the 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
منابع مشابه
Supplementary Material: Deep Adaptive Image Clustering
This is the supplementary material for the paper entitled “Deep Adaptive Image Clustering”. The supplementary material is organized as follows. Section 1 gives the mapping function described in Figure 1. Section 2 presents the proof of Theorem 1. Section 3 details the experimental settings in our experiments. 1. The Mapping Function Utilized in Figure 1 We assume that li represents the label fe...
متن کاملAggregating Image and Text Quantized Correlated Components Supplementary Material
This is the supplementary material to the article Aggregating Image and Text Quantized Correlated Components published at CVPR 2016. While not necessary to understand the work exposed in the paper, it reports some complementary results or “interesting negative results”. Last section focuses on providing implementation details to reproduce the experiments, in particular regarding the material co...
متن کاملLearning a Dictionary of Shape Epitomes with Applications to Image Labeling: Supplementary Material
In the main paper, we have proposed a new representation for image shape: a dictionary of shape epitomes, which explicitly include hidden variables to encode shift and rotation. We have also applied this representation for semantic image labeling. In this supplementary material, we further show the versatility of shape epitomes by employing them for edge detection and local appearance modeling.
متن کاملCombining Geometric, Textual and Visual Features for Predicting Prepositions in Image Descriptions: Supplementary Material
Prepositions in Image Descriptions: Supplementary Material Arnau Ramisa* Josiah Wang* Ying Lu Emmanuel Dellandrea Francesc Moreno-Noguer Robert Gaizauskas 1 Institut de Robòtica i Informàtica Industrial (UPC-CSIC), Barcelona, Spain 2 Department of Computer Science, University of Sheffield, UK 3 LIRIS, École Centrale de Lyon, France {aramisa, fmoreno}@iri.upc.edu {j.k.wang, r.gaizauskas}@sheffie...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2015