An original approach for the localization of objects in images

By R. Egis Vaillant, C. Monrocq, Y. Le Cun. 1993

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

OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks (2013, background): “ and size. Many authors have proposed to use ConvNets for detection and localization with a sliding window over multiple scales, going back to the early 1990’s for multi-character strings [20], faces [30], and hands [22]. More recently, ConvNets have been shown to yield state of the art performance ”
Feature Pyramid Networks for Object Detection (2016, background): “6] demonstrated fast pyramid computation by ﬁrst computing a sparsely sampled (in scale) pyramid and then interpolating missing levels. Before HOG and SIFT, early work on face detection with ConvNets [38,32] computed shallow networks over image pyramids to detect faces across scales. 2 Deep ConvNet ”
Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks (2015, background): “, but, in particular, for small objects. 2. Prior work ConvNet object detectors. ConvNets with a small number of hidden layers have been used for object detection for the last two decades (e.g., from [39] to [34]). Until recently, they were successful in restricted domains such as face detection. Re”
Deformable part models are convolutional neural networks (2014, methodology): “andard technique for coping with this fact is to run a detector at multiple scales using an image pyramid. In the context of CNNs, this method dates back to (at least) early work on face detection in [31], and has been used in contemporary work including OverFeat [28], DetectorNet [3], and DenseNet ”
Deep learning in neural networks: An overview (2014, methodology): “t computations. Recent MPCNN scanners speed up naive implementations by up to three orders of magnitude(Masci et al., 2013; Giusti et al., 2013); compare earlier efﬁcient methods for CNNs without MP (Vaillant et al., 1994). Also in 2012, a system consisting of growing deep FNNs and 2D-BRNNs (Di Lena”
DenseNet: Implementing Efficient ConvNet Descriptor Pyramids (2014, neutral): “et al. compute CNN pyramids and perform sliding-window processing for image segmentation [18]. Several years earlier, Vaillant et al. densely computed CNNs on full images for robust face localization [19]. 3 DenseNet CNN Feature Pyramids The CNN object classier of [4] operates on xed size square ima”
Convolutional face finder: a neural architecture for fast and robust face detection (2004, methodology): “[33], input images of size 20 20 are preprocessed with a Laplacian filter, normalized and passed to an architecture composed of one convolutional layer of four feature maps of size 16 16, followed by a subsampling layer of four feature maps.”
Actions and Attributes from Wholes and Parts (2014, background): “This is one of the oldest “tricks” in computer vision and has been implemented by slidingwindow object detection approaches throughout the years [1, 9, 22, 27].”
TensorMask: A Foundation for Dense Object Segmentation (2019, background): “ 1. Introduction The sliding-window paradigm—ﬁnding objects by looking in each window placed over a dense set of image locations—is one of the earliest and most successful concepts in computer vision [37,39,9,10] and is naturally connected to convolutional networks [20]. However, while today’s toppe”
Focal Loss for Dense Object Detection (2017, background): “pplied on a dense image grid, has a long and rich history. One of the earliest successes is the classic work of LeCun et al. who applied convolutional neural networks to handwritten digit recognition [19,36]. Viola and Jones [37] used boosted object detectors for face detection, leading to widesprea”
FindIt: Generalized Localization with Natural Language Queries (2022, background): “A highly related semantic localization task is object detection (DET), which seeks to detect all objects from a predefined set of classes without text inputs [60,71,77,80,18,68].”