Learning to detect objects in images via a sparse, part-based representation

By Shivani Agarwal, Aatif Awan, Dan Roth. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004

1,001 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/b82d251ed367593366680acebc81fdb070b04a18

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

UvA-DARE ( Digital Academic Repository ) The visual extent of an object : suppose we know the object locations (2011, methodology): “The first step on the road to less localisation of the object was to use local region descriptors in a specific spatial arrangement (Agarwal et al. 2004; Burl et al. 1998; Fergus et al. 2003).”
Efficient Subwindow Search: A Branch and Bound Framework for Object Localization (2009, neutral): “We use the data set in its original setup [34] where the task is pure localization.”
Graph Regularized Sparse Coding for Image Representation (2011, background): “It has been proven useful in many applications [1]–[3].”
A Review of Computer Vision Techniques for the Analysis of Urban Traffic (2011, methodology): “A similar approach is used in [2]; however, the relations”
Merging chrominance and luminance in early, medium, and late fusion using Convolutional Neural Networks (2015, methodology): “The dataset was used in two experiments reported in [Agarwal et al., 2004; Agarwal and Roth, 2002].”
Robust Fragments-based Tracking using the Integral Histogram (2006, neutral): “We present extensive experimental results on challenging sequences, which demonstrate the robust tracking achieved by our algorithm (even with the use of only gray-scale (noncolor) information).”
The 2005 PASCAL Visual Object Classes Challenge (2005, methodology): “In recent years there has been a rapid growth in research, and quite some success, in visual recognition of object classes; examples include [1, 5, 10, 14, 18, 28, 39, 43].”
Robust Object Detection with Interleaved Categorization and Segmentation (2008, methodology): “In this paper, we compare simple Greyvalue Patches (Agarwal et al. 2004), SIFT (Lowe 2004), and Local Shape Context (Belongie et al. 2002; Mikolajczyk and Schmid 2005) descriptors.”
Hough-based tracking of non-rigid objects (2011, methodology): “To alleviate the influence of occlusions to the overall performance, we measure the tracking accuracy using the Agarwal-criterion [1], which is defined as score = RT ∩RGT RT , where R T is the tracking rectangle and R GT the ground truth.”
MoSIFT: Recognizing Human Actions in Surveillance Videos (2009, background): “This part-based approach [16] assumes that a collection of distinctive parts can effectively describe the whole object.”
The segmented and annotated IAPR TC-12 benchmark (2010, background): “Most of these, however, are restricted to speciﬁc domains, such as cars [26], nature-roadways [27], animals [28], landscape vs. structured classiﬁcation [29], and natural scene classiﬁcation [30].”
Fast PRISM: Branch and Bound Hough Transform for Object Class Detection (2011, methodology): “(2009), we use the UIUC cars database (Agarwal et al. 2004) and focus on the multi-scale subset.”
Beyond sliding windows: Object localization by efficient subwindow search (2008, methodology): “We use the dataset in its original setup [1] where the task is pure localization.”
Utility data annotation with Amazon Mechanical Turk (2008, background): “It is obvious from this figure that hips are much more difficult to localize compared to shoulders, knees, elbows, wrists, ankles, the head and the neck.”
The Visual Extent of an Object (2011, methodology): “The first step on the road to less localisation of the object was to use local region descriptors in a specific spatial arrangement (Agarwal et al. 2004; Burl et al. 1998; Fergus et al. 2003).”
Efficient Segmentation: Learning Downsampling Near Semantic Boundaries (2019, background): “Computational requirements of the contemporaneous approaches and the cost of annotations conditioned the low resolution of images or imprecise (rough) annotations in popular semantic segmentation datasets, such as Caltech [15], [3], Pascal VOC [13, 14, 20], COCO [34].”
Class-specific Hough forests for object detection (2009, background): “The UIUC car dataset [1] contains images of side views of cars.”
Part-dependent Label Noise: Towards Instance-dependent Label Noise (2020, background): “The rational behind this assumption is that the learned parts are semantic [29], and their contributions to perceiving the instance should be similar in the contributions to understanding and annotating them [7, 3].”