• Corpus ID: 15729146

Boosting Convolutional Features for Robust Object Proposals

@article{Karianakis2015BoostingCF,
  title={Boosting Convolutional Features for Robust Object Proposals},
  author={Nikolaos Karianakis and Thomas J. Fuchs and Stefano Soatto},
  journal={ArXiv},
  year={2015},
  volume={abs/1503.06350}
}
Deep Convolutional Neural Networks (CNNs) have demonstrated excellent performance in image classification, but still show room for improvement in object-detection tasks with many categories, in particular for cluttered scenes and occlusion. Modern detection algorithms like Regions with CNNs (Girshick et al., 2014) rely on Selective Search (Uijlings et al., 2013) to propose regions which with high probability represent objects, where in turn CNNs are deployed for classification. Selective Search… 

Figures and Tables from this paper

Object Proposal Generation With Fully Convolutional Networks
TLDR
This paper presents a framework utilizing fully convolutional networks (FCNs) to produce object proposal positions and bounding box location refinement with Support Vector Machine (SVM) to further improve proposal localization.
Convolutional Channel Features For Pedestrian, Face and Edge Detection
TLDR
Convolutional Channel Features (CCF), which transfers low-level features from off-the-shelf ConvNet models to feed the boosting classifiers based on decision trees, benefits from the rich capacity, robustness and sparsity in feature representation, as well as more efficiency in computation and storage during inference and learning process.
Convolutional Channel Features
TLDR
Convolutional Channel Features (CCF) serves as a good way of tailoring pre-trained CNN models to diverse tasks without fine-tuning the whole network to each task by achieving state-of-the-art performances in pedestrian detection, face detection, edge detection and object proposal generation.
Exploit All the Layers: Fast and Accurate CNN Object Detector with Scale Dependent Pooling and Cascaded Rejection Classifiers
In this paper, we investigate two new strategies to detect objects accurately and efficiently using deep convolutional neural network: 1) scale-dependent pooling and 2) layerwise cascaded rejection
Cascaded Sparse Spatial Bins for Efficient and Effective Generic Object Detection
TLDR
A novel efficient method for extraction of object proposals that exploits deep spatial pyramid features, a novel fast-to-compute HoG-based edge statistic and the EdgeBoxes score to achieve state-of-the-art recall performance on Pascal VOC07.
Convolutional Channel Features: Tailoring CNN to Diverse Tasks
TLDR
Convolutional Channel Features (CCF) serves as a good way of tailoring pre-trained CNN models to diverse tasks without fine-tuning the whole network to each task by achieving state-of-the-art performances in pedestrian detection, face detection, edge detection and object proposal generation.
Texture Complexity Based Redundant Regions Ranking for Object Proposal
TLDR
Experimental results on the PASCAL VOC 2007 dataset show that the proposed TCR significantly improves the baseline approach by increasing AUC (area under recall curve) and outperforms the state-of-the-art with AUC and uses fewer detection proposals to achieve comparable recall rates.
Multi-scale energy optimization for object proposal generation
TLDR
An object proposal generation method by applying energy optimization into superpixel merging algorithms in a multiscale framework, which could generate possible object locations in one image by taking advantage of robust edge detection and segments each image only once, which greatly reduces the number of proposals.
Sampling Algorithms to Handle Nuisances in Large-Scale Recognition
TLDR
DSP-CNN is introduced, which deploys domain-size pooling in order to transform the neural networks to be scale invariant in the convolutional operator level, and novel sampling and pruning techniques for region proposal schemes that improve the end-to-end performance in large-scale classification, detection and wide-baseline correspondence to state-of-the-art levels are proposed.
Learning to Count with CNN Boosting
TLDR
This paper employs CNNs and incorporates two significant improvements to the state of the art methods: layered boosting and selective sampling to increase the counting accuracy and to reduce processing time.
...
...

References

SHOWING 1-10 OF 41 REFERENCES
Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
TLDR
This paper proposes a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%.
Scalable Object Detection Using Deep Neural Networks
TLDR
This work proposes a saliency-inspired neural network model for detection, which predicts a set of class-agnostic bounding boxes along with a single score for each box, corresponding to its likelihood of containing any object of interest.
Scalable, High-Quality Object Detection
TLDR
It is demonstrated that learning-based proposal methods can effectively match the performance of hand-engineered methods while allowing for very efficient runtime-quality trade-offs.
Regionlets for Generic Object Detection
TLDR
This work proposes to model an object class by a cascaded boosting classifier which integrates various types of features from competing local regions, named as region lets, which significantly outperforms the state-of-the-art on popular multi-class detection benchmark datasets with a single method.
What Makes for Effective Detection Proposals?
TLDR
An in-depth analysis of twelve proposal methods along with four baselines regarding proposal repeatability, ground truth annotation recall on PASCAL, ImageNet, and MS COCO, and their impact on DPM, R-CNN, and Fast R- CNN detection performance shows that for object detection improving proposal localisation accuracy is as important as improving recall.
Return of the Devil in the Details: Delving Deep into Convolutional Nets
TLDR
It is shown that the data augmentation techniques commonly applied to CNN-based methods can also be applied to shallow methods, and result in an analogous performance boost, and it is identified that the dimensionality of the CNN output layer can be reduced significantly without having an adverse effect on performance.
Deep Joint Task Learning for Generic Object Extraction
TLDR
This paper investigates how to extract objects-of-interest without relying on handcraft features and sliding windows approaches, and presents a general joint task learning framework, in which each task is tackled via a multi-layer convolutional neural network, and the two networks work collaboratively to boost performance.
Learning a category independent object detection cascade
TLDR
This work focuses on the first layers of a category independent object detection cascade in which a large number of windows from an objectness prior are sampled, and then discriminatively learn to filter these candidate windows by an order of magnitude.
Very Deep Convolutional Networks for Large-Scale Image Recognition
TLDR
This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
TLDR
This work equips the networks with another pooling strategy, “spatial pyramid pooling”, to eliminate the above requirement, and develops a new network structure, called SPP-net, which can generate a fixed-length representation regardless of image size/scale.
...
...