What, where and who? Classifying events by scene and object recognition

By Li-Jia Li, Li Fei-Fei. 2007

864 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

NAS evaluation is frustratingly hard (2019, methodology): “…to the standard CIFAR10 we select CIFAR100 for a more challenging object classiﬁcation problem (Krizhevsky, 2009); SPORT8 for action classiﬁcation (Li & Fei-Fei, 2007); MIT67 for scene classiﬁcation (Quattoni & Torralba, 2009); and FLOWERS102 for ﬁne-grained object classiﬁcation (Nilsback &…”
Interpreting Deep Visual Representations via Network Dissection (2017, neutral): “We include the event8 [51], action40 [52], indoor67 [53], sun397 [54], caltech101 [55], and caltech256 [56].”
Neural Motifs: Scene Graph Parsing with Global Context (2017, methodology): “Our approach is most closely related to work that models object co-occurrence using graphical models to combine many sources of contextual information [30, 11, 23, 10].”
An empirical study of context in object detection (2009, background): “Semantic context might indicate the kind of event, activity, or other scene category being depicted [1, 22 , 28].”
See No Evil, Say No Evil: Description Generation from Densely Labeled Images (2014, background): “Activity recognition is an emerging area in images (Li and Fei-Fei, 2007; Yao et al., 2011; Sharma et al., 2013) and video (Weinland et al., 2011), although less studied than object recognition.”
Asynchronous Temporal Fields for Action Recognition (2016, background): “Understanding activities with emphasis on objects has been a particularly fruitful direction [25, 36, 9, 34, 54].”
Monocular human pose estimation: A survey of deep learning-based methods (2020, background): “…Ferrari, 2012b) (sports, full-body image, upper body annotation). full-body pose datasets include PASCAL Person Layout (Everingham et al., 2010) (daily scene, from PASCAL VOC (Everingham et al., 2010)), Sport (Wang et al., 2011) (sport scenes) and UIUC people (Li and Fei-fei, 2007) (sport scenes).”
Hierarchical Subquery Evaluation for Active Learning on a Graph (2014, neutral): “UIUCSport [21] 1579 50 8 PCA BoW 0.”
Kernel Sparse Representation for Image Classification and Face Recognition (2010, background): “UIUC-Sport [15] contains images collected from 8 kind of different sports: badminton, bocce,croquet, polo, rock climbing, rowing, sailing and snow boarding.”
Exemplar based Deep Discriminative and Shareable Feature Learning for scene image classification (2015, methodology): “less performance gain. 5. Experiments and Analysis 5.1. Datasets and Experiment Settings We tested our DDSFL method on three widely used scene image classication datasets: Scene 15 [11], UIUC Sports [53], and MIT Indoor [54]. We also tested on the challenging PASCAL VOC 2012 object classication data”
Object Bank: A High-Level Image Representation for Scene Classification & Semantic Feature Sparsification (2010, methodology): “70 randomly drawn images from each classes are used for training and 60 for testing following [22].”
A Deep and Autoregressive Approach for Topic Modeling of Multimodal Data (2014, methodology): “We tested our model on 2 real-world data sets: a subset of the LabelMe data set [29] and the UIUC-Sports data set [30].”
Max-margin Latent Dirichlet Allocation for Image Classification and Annotation (2011, methodology): “Much work [1, 3, 4, 12, 13, 20, 21] in image classification and labeling uses topic models, which are a class of powerful tools originally proposed in text modeling and have gained much popularity in computer vision recently.”
Segmentation and Recognition Using Structure from Motion Point Clouds (2008, methodology): “Many successful techniques for 2D object recognition have used individual still images [6–8].”