2018

AutoFocus: Efficient Multi-Scale Inference

Najibi, Mahyar, Singh, Bharat, Davis, Larry S.

Understand

This paper describes AutoFocus, an efficient multi-scale inference algorithm for deep-learning based object detectors.

Instead of processing an entire image pyramid, AutoFocus adopts a coarse to fine approach and only processes regions which are likely to contain small objects at finer scales.
This is achieved by predicting category agnostic segmentation maps for small objects at coarser scales, called FocusPixels.
FocusPixels can be predicted with high recall, and in many cases, they only cover a small fraction of the entire image.

Built on

Saccadic suppression: a review and an analysis
E. Matin · 1974
Earlier work this paper cites.
Implications of sustained and transient channels for theories of visual pattern masking, saccadic suppression, and information processing
B. G. Breitmeyer and L. Ganz · 1976
Earlier work this paper cites.
Theory of edge detection
D. Marr and E. Hildreth · 1980
Earlier work this paper cites.
Scale-space filtering: A new approach to multi-scale description
A. Witkin · 1984
Earlier work this paper cites.
The laplacian pyramid as a compact image code
P. J. Burt and E. H. Adelson · 1987
Earlier work this paper cites.
Active vision
J. Aloimonos, I. Weiss, and A. Bandyopadhyay · 1988
Earlier work this paper cites.
Visual masking and visual integration across saccadic eye movements
D. E. Irwin, J. S. Brown, and J.-s. Sun · 1988
Earlier work this paper cites.
Simple anatomy of the retina
H. Kolb · 1995
Earlier work this paper cites.
A model of saliency-based visual attention for rapid scene analysis
L. Itti, C. Koch, and E. Niebur · 1998
Earlier work this paper cites.
Gradient-based learning applied to document recognition
Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner · 1998
Earlier work this paper cites.
Rapid object detection using a boosted cascade of simple features
P. Viola and M. Jones · 2001
Earlier work this paper cites.
Active vision: The psychology of looking and seeing
J. M. Findlay and I. D. Gilchrist · 2003
Earlier work this paper cites.
Distinctive image features from scale-invariant keypoints
D. G. Lowe · 2004
Earlier work this paper cites.
Scale & affine invariant interest point detectors
K. Mikolajczyk and C. Schmid · 2004
Earlier work this paper cites.
Robust object detection via soft cascade
L. Bourdev and J. Brandt · 2005
Earlier work this paper cites.
Histograms of oriented gradients for human detection
N. Dalal and B. Triggs · 2005
Earlier work this paper cites.
Detecting pedestrians using patterns of motion and appearance
P. Viola, M. J. Jones, and D. Snow · 2005
Earlier work this paper cites.
Surf: Speeded up robust features
H. Bay, T. Tuytelaars, and L. Van Gool · 2006
Earlier work this paper cites.
Fast human detection using a cascade of histograms of oriented gradients
Q. Zhu, M.-C. Yeh, K.-T. Cheng, and S. Avidan · 2006
Earlier work this paper cites.
Saliency detection: A spectral residual approach
X. Hou and L. Zhang · 2007
Earlier work this paper cites.
Multiple-instance pruning for learning efficient cascade detectors
C. Zhang and P. A. Viola · 2008
Earlier work this paper cites.
Integral channel features
P. Dollár, Z. Tu, P. Perona, and S. Belongie · 2009
Earlier work this paper cites.
Center-surround patterns emerge as optimal predictors for human saccade targets
W. Kienzle, M. O. Franz, B. Schölkopf, and F. A. Wichmann · 2009
Earlier work this paper cites.
Constrained parametric min-cuts for automatic object segmentation
J. Carreira and C. Sminchisescu · 2010
Earlier work this paper cites.
The fastest pedestrian detector in the west
P. Dollár, S. J. Belongie, and P. Perona · 2010
Earlier work this paper cites.

Similar

The pascal visual object classes (voc) challenge
M. Everingham, L. Van Gool, C. K. Williams, J. Winn, and A. Zisserman · 2010
Cited alongside, same era.
Object detection with discriminatively trained part-based models
P. F. Felzenszwalb, R. B. Girshick, D. McAllester, and D. Ramanan · 2010
Cited alongside, same era.
Learning to detect a salient object
T. Liu, Z. Yuan, J. Sun, J. Wang, N. Zheng, X. Tang, and H.-Y. Shum · 2011
Cited alongside, same era.
Peripheral vision and pattern recognition: A review
H. Strasburger, I. Rentschler, and M. Jüttner · 2011
Cited alongside, same era.
Pedestrian detection at 100 frames per second
R. Benenson, M. Mathias, R. Timofte, and L. Van Gool · 2012
Cited alongside, same era.
Crosstalk cascades for frame-rate pedestrian detection
P. Dollár, R. Appel, and W. Kienzle · 2012
Cited alongside, same era.
Selective search for object recognition
J. R. Uijlings, K. E. Van De Sande, T. Gevers, and A. W. Smeulders · 2013
Cited alongside, same era.
Fast feature pyramids for object detection
P. Dollár, R. Appel, S. Belongie, and P. Perona · 2014
Cited alongside, same era.
Microsoft coco: Common objects in context
T.-Y. Lin, M. Maire, S. Belongie, J. Hays, P. Perona, D. Ramanan, P. Dollár, and C. L. Zitnick · 2014
Cited alongside, same era.
An active search strategy for efficient object class detection
A. Gonzalez-Garcia, A. Vezhnevets, and V. Ferrari · 2015
Cited alongside, same era.
Fully convolutional networks for semantic segmentation
J. Long, E. Shelhamer, and T. Darrell · 2015
Cited alongside, same era.
Faster r-cnn: Towards real-time object detection with region proposal networks
S. Ren, K. He, R. Girshick, and J. Sun · 2015
Cited alongside, same era.

Then

Light-head r-cnn: In defense of two-stage object detector
Original
Z. Li, C. Peng, G. Yu, X. Zhang, Y. Deng, and J. Sun · 2017
Later among the works it cites.
Feature pyramid networks for object detection
T.-Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan, and S. Belongie · 2017
Later among the works it cites.
Recurrent scale approximation for object detection in cnn
Y. Liu, H. Li, J. Yan, F. Wei, X. Wang, and X. Tang · 2017
Later among the works it cites.
SSH: Single stage headless face detector
M. Najibi, P. Samangouei, R. Chellappa, and L. Davis · 2017
Later among the works it cites.
Yolo9000: Better, faster, stronger
J. Redmon and A. Farhadi · 2017
Later among the works it cites.
Fast yolo: A fast you only look once system for real-time embedded object detection in video
Original
M. J. Shafiee, B. Chywl, F. Li, and A. Wong · 2017
Later among the works it cites.
Aggregated residual transformations for deep neural networks
S. Xie, R. Girshick, P. Dollár, Z. Tu, and K. He · 2017
Later among the works it cites.
Cascade r-cnn: Delving into high quality object detection
Z. Cai and N. Vasconcelos · 2018
Closest in time.
Optimizing video object detection via a scale-time lattice
K. Chen, J. Wang, S. Yang, X. Zhang, Y. Xiong, C. Change Loy, and D. Lin · 2018
Closest in time.
Dynamic zoom-in network for fast object detection in large images
M. Gao, R. Yu, A. Li, V. I. Morariu, and L. S. Davis · 2018
Closest in time.
Cornernet: Detecting objects as paired keypoints
H. Law and J. Deng · 2018
Closest in time.
Focal loss for dense object detection
T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár · 2018
Closest in time.
Dynamic deep neural networks: Optimizing accuracy-efficiency trade-offs by selective execution
L. Liu and J. Deng · 2018
Closest in time.
Mobile video object detection with temporally-aware feature maps
M. Liu and M. Zhu · 2018
Closest in time.
Path aggregation network for instance segmentation
S. Liu, L. Qi, H. Qin, J. Shi, and J. Jia · 2018
Closest in time.
Towards the success rate of one: Real-time unconstrained salient object detection
M. Najibi, F. Yang, Q. Wang, and R. Piramuthu · 2018
Closest in time.
Megdet: A large mini-batch object detector
C. Peng, T. Xiao, Z. Li, Y. Jiang, X. Zhang, K. Jia, G. Yu, and J. Sun · 2018
Closest in time.
Deep reinforcement learning of region proposal networks for object detection
A. Pirinen and C. Sminchisescu · 2018
Closest in time.
An analysis of scale invariance in object detection-snip
B. Singh and L. S. Davis · 2018
Closest in time.
SNIPER: Efficient multi-scale training
B. Singh, M. Najibi, and L. S. Davis · 2018
Closest in time.
Beyond trade-off: Accelerate fcn-based face detector with higher accuracy
G. Song, Y. Liu, M. Jiang, Y. Wang, J. Yan, and B. Leng · 2018
Closest in time.
Blockdrop: Dynamic inference paths in residual networks
Z. Wu, T. Nagarajan, A. Kumar, S. Rennie, L. S. Davis, K. Grauman, and R. Feris · 2018
Closest in time.
Single-shot refinement neural network for object detection
S. Zhang, L. Wen, X. Bian, Z. Lei, and S. Z. Li · 2018
Closest in time.
Deformable convnets v2: More deformable, better results
Original
X. Zhu, H. Hu, S. Lin, and J. Dai · 2018
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…