Multiresolution Models for Object Detection

  title={Multiresolution Models for Object Detection},
  author={Dennis Park and Deva Ramanan and Charless C. Fowlkes},
  booktitle={European Conference on Computer Vision},
Most current approaches to recognition aim to be scale-invariant. However, the cues available for recognizing a 300 pixel tall object are qualitatively different from those for recognizing a 3 pixel tall object. We argue that for sensors with finite resolution, one should instead use scale-variant, or multiresolution representations that adapt in complexity to the size of a putative detection window. We describe a multiresolution model that acts as a deformable part-based model when scoring… 

Detection and localization with multi-scale models

Experimental analysis on the PASCAL VOC dataset shows the method to considerably improve both detection and localization performance for different type of features, histogram of oriented gradients and deep convolutional neural network features.

Discriminative Models for Multi-Class Object Layout

A unified model for multi-class object recognition is introduced that casts the problem as a structured prediction task and how to formulate learning as a convex optimization problem is shown.

Fast Feature Pyramids for Object Detection

For a broad family of features, this work finds that features computed at octave-spaced scale intervals are sufficient to approximate features on a finely-sampled pyramid, and this approximation yields considerable speedups with negligible loss in detection accuracy.

Data-driven exemplar model selection

This work proposes a two stage data-driven process, which selects and learns a compact set of exemplar models for object detection, which has an inherent ranking, which can be used for anytime/budgeted detection scenarios.

Seeking the Strongest Rigid Detector

This paper revisits some of the core assumptions in HOG+SVM and shows that by properly designing the feature pooling, feature selection, preprocessing, and training methods, it is possible to reach top quality, at least for pedestrian detections, using a single rigid component.

Multi-scale volumes for deep object detection and localization

Finding Tiny Faces

  • Peiyun HuD. Ramanan
  • Computer Science
    2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2017
The role of scale in pre-trained deep networks is explored, providing ways to extrapolate networks tuned for limited scales to rather extreme ranges and demonstrating state-of-the-art results on massively-benchmarked face datasets.

Object Detection with Appearance-based Mixture Models Anonymous CVPR submission

  • Computer Science
  • 2010
A simple mixture model for object detection based on unsupervised appearance-based clustering of object instances is presented, requiring no additional labels or heuristics, and it yields improved performance in the detection task.

Scalable multi-class object detection

This work presents a scalable multi-class detection algorithm which scales sublinearly with the number of classes without compromising accuracy, and builds a taxonomy of object classes which is exploited to further reduce the cost of multi- class object detection.

Recognizing and Segmenting Objects in the Presence of Occlusion and Clutter

A deformable part model for detection and keypoint localization is introduced that explicitly models part occlusion and this approach yields state-of-the-art semantic segmentation results without resorting to morecomplex random-field inference or instance detection driven architectures.



Discriminative Models for Multi-Class Object Layout

A unified model for multi-class object recognition is introduced that casts the problem as a structured prediction task and how to formulate learning as a convex optimization problem is shown.

Multiple Component Learning for Object Detection

The method, Multiple Component Learning (mcl), automatically learns individual component classifiers and combines these into an overall classifier, and unlike methods that are not part-based, mcl is quite robust to occlusions.

Example-Based Object Detection in Images by Components

Results suggest that the improvement in performance is due to the component-based approach and the ACC data classification architecture, which is capable of locating partially occluded views of people and people whose body parts have little contrast with the background.

Object class recognition by unsupervised scale-invariant learning

The flexible nature of the model is demonstrated by excellent results over a range of datasets including geometrically constrained classes (e.g. faces, cars) and flexible objects (such as animals).

A discriminatively trained, multiscale, deformable part model

A discriminatively trained, multiscale, deformable part model for object detection, which achieves a two-fold improvement in average precision over the best performance in the 2006 PASCAL person detection challenge and outperforms the best results in the 2007 challenge in ten out of twenty categories.

Detecting Pedestrians by Learning Shapelet Features

This paper introduces an algorithm for learning shapelet features, a set of mid-level features that are built from low-level gradient information that discriminates between pedestrian and non-pedestrian classes on the INRIA dataset.

Multiple instance fFeature for robust part-based object detection

This work proposes a learning framework to mitigate feature misalignment in object detection, where a boosting algorithm is performed to seed the position of the object part, and a multiple instance boosting algorithm further pursues an aggregated feature for this part, namely multiple instance feature.

Pedestrian detection using wavelet templates

This paper presents a trainable object detection architecture that is applied to detecting people in static images of cluttered scenes and shows how the invariant properties and computational efficiency of the wavelet template make it an effective tool for object detection.

Pedestrian detection: A benchmark

The Caltech Pedestrian Dataset is introduced, which is two orders of magnitude larger than existing datasets and proposes improved evaluation metrics, demonstrating that commonly used per-window measures are flawed and can fail to predict performance on full images.

Histograms of oriented gradients for human detection

  • N. DalalB. Triggs
  • Computer Science
    2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05)
  • 2005
It is shown experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection, and the influence of each stage of the computation on performance is studied.