object detection, deep learning, and r-cnns · object detection, deep learning, and r-cnns partly...
TRANSCRIPT
![Page 1: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/1.jpg)
Object detection, deep learning, and R-CNNs
Partly from Ross Girshick
Microsoft Research
Now at Facebook
![Page 2: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/2.jpg)
Pedestrians
Histograms of Oriented Gradients for Human Detection, Dalal and Triggs, CVPR 2005
AP ~77%More sophisticated methods: AP ~90%
(a) average gradient image over training examples(b) each “pixel” shows max positive SVM weight in the block centered on that pixel(c) same as (b) for negative SVM weights(d) test image(e) its R-HOG descriptor(f) R-HOG descriptor weighted by positive SVM weights(g) R-HOG descriptor weighted by negative SVM weights
![Page 3: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/3.jpg)
Overview of HOG Method
1. Compute gradients in the region to be described
2. Put them in bins according to orientation
3. Group the cells into large blocks
4. Normalize each block
5. Train classifiers to decide if these are parts of a human
![Page 4: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/4.jpg)
Details
• Gradients[-1 0 1] and [-1 0 1]T were good enough filters.
• Cell HistogramsEach pixel within the cell casts a weighted vote for an orientation-based histogram channel based on the values found in the gradient computation. (9 channels worked)
• BlocksGroup the cells together into larger blocks, either R-HOGblocks (rectangular) or C-HOG blocks (circular).
![Page 5: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/5.jpg)
More Details
• Block Normalization
They tried 4 different kinds of normalization.Let be the block to be normalized and e be a small constant.
![Page 6: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/6.jpg)
Example: Dalal-Triggs pedestrian detector
1. Extract fixed-sized (64x128 pixel) window at each position and scale
2. Compute HOG (histogram of gradient) features within each window
3. Score the window with a linear SVM classifier
4. Perform non-maxima suppression to remove overlapping detections with lower scores
Navneet Dalal and Bill Triggs, Histograms of Oriented Gradients for Human Detection, CVPR05
![Page 7: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/7.jpg)
Slides by Pete Barnum Navneet Dalal and Bill Triggs, Histograms of Oriented Gradients for Human Detection, CVPR05
![Page 8: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/8.jpg)
uncentered
centered
cubic-corrected
diagonal
Sobel
Slides by Pete Barnum Navneet Dalal and Bill Triggs, Histograms of Oriented Gradients for Human Detection, CVPR05
Outperforms
![Page 9: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/9.jpg)
• Histogram of gradient orientations
• Votes weighted by magnitude
• Bilinear interpolation between cells
Orientation: 9 bins (for unsigned angles)
Histograms in 8x8 pixel cells
Slides by Pete Barnum Navneet Dalal and Bill Triggs, Histograms of Oriented Gradients for Human Detection, CVPR05
![Page 10: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/10.jpg)
Normalize with respect to surrounding cells
Slides by Pete Barnum Navneet Dalal and Bill Triggs, Histograms of Oriented Gradients for Human Detection, CVPR05
![Page 11: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/11.jpg)
X=
Slides by Pete Barnum Navneet Dalal and Bill Triggs, Histograms of Oriented Gradients for Human Detection, CVPR05
# features = 15 x 7 x 9 x 4 = 3780
# cells
# orientations
# normalizations by neighboring cells
![Page 12: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/12.jpg)
Training set
![Page 13: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/13.jpg)
Slides by Pete Barnum Navneet Dalal and Bill Triggs, Histograms of Oriented Gradients for Human Detection, CVPR05
pos w neg w
![Page 14: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/14.jpg)
pedestrian
Slides by Pete Barnum Navneet Dalal and Bill Triggs, Histograms of Oriented Gradients for Human Detection, CVPR05
![Page 15: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/15.jpg)
Detection examples
![Page 16: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/16.jpg)
![Page 17: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/17.jpg)
Deformable Parts Model
• Takes the idea a little further
• Instead of one rigid HOG model, we have multiple HOG models in a spatial arrangement
• One root part to find first and multiple other parts in a tree structure.
![Page 18: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/18.jpg)
The Idea
Articulated parts model• Object is configuration of parts
• Each part is detectable
Images from Felzenszwalb
![Page 19: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/19.jpg)
Deformable objects
Images from Caltech-256
Slide Credit: Duan Tran
![Page 20: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/20.jpg)
Deformable objects
Images from D. Ramanan’s datasetSlide Credit: Duan Tran
![Page 21: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/21.jpg)
How to model spatial relations?• Tree-shaped model
![Page 22: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/22.jpg)
![Page 23: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/23.jpg)
Hybrid template/parts model
Detections
Template Visualization
Felzenszwalb et al. 2008
![Page 24: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/24.jpg)
Pictorial Structures Model
Appearance likelihood Geometry likelihood
![Page 25: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/25.jpg)
Results for person matching
27
![Page 26: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/26.jpg)
Results for person matching
28
![Page 27: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/27.jpg)
BMVC 2009
![Page 28: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/28.jpg)
2012 State-of-the-art Detector:Deformable Parts Model (DPM)
30Felzenszwalb et al., 2008, 2010, 2011, 2012
Lifetime
Achievement
1. Strong low-level features based on HOG2. Efficient matching algorithms for deformable part-based
models (pictorial structures)3. Discriminative learning with latent variables (latent SVM)
![Page 29: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/29.jpg)
Why did gradient-based models work?
Average gradient image
![Page 30: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/30.jpg)
Generic categories
Can we detect people, chairs, horses, cars, dogs, buses, bottles, sheep …?PASCAL Visual Object Categories (VOC) dataset
![Page 31: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/31.jpg)
Generic categoriesWhy doesn’t this work (as well)?
Can we detect people, chairs, horses, cars, dogs, buses, bottles, sheep …?PASCAL Visual Object Categories (VOC) dataset
![Page 32: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/32.jpg)
PASCAL VOC detection history
0%
10%
20%
30%
40%
50%
60%
70%
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
mea
n A
vera
ge P
reci
sio
n (
mA
P)
year
DPM
DPM,HOG+BOW
DPM,MKL
DPM++DPM++,MKL,SelectiveSearch
Selective Search,DPM++,MKL
41%41%37%
28%23%
17%
![Page 33: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/33.jpg)
Part-based models & multiple features (MKL)
0%
10%
20%
30%
40%
50%
60%
70%
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
mea
n A
vera
ge P
reci
sio
n (
mA
P)
year
DPM
DPM,HOG+BOW
DPM,MKL
DPM++DPM++,MKL,SelectiveSearch
Selective Search,DPM++,MKL
41%41%37%
28%23%
17%
![Page 34: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/34.jpg)
Kitchen-sink approaches
0%
10%
20%
30%
40%
50%
60%
70%
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
mea
n A
vera
ge P
reci
sio
n (
mA
P)
year
DPM
DPM,HOG+BOW
DPM,MKL
DPM++DPM++,MKL,SelectiveSearch
Selective Search,DPM++,MKL
41%41%37%
28%23%
17%
increasing complexity & plateau
![Page 35: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/35.jpg)
Region-based Convolutional Networks (R-CNNs)
0%
10%
20%
30%
40%
50%
60%
70%
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
mea
n A
vera
ge P
reci
sio
n (
mA
P)
year
DPM
DPM,HOG+BOW
DPM,MKL
DPM++DPM++,MKL,SelectiveSearch
Selective Search,DPM++,MKL
41%41%37%
28%23%
17%
53%
62%
R-CNN v1
R-CNN v2
[R-CNN. Girshick et al. CVPR 2014]
![Page 36: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/36.jpg)
0%
10%
20%
30%
40%
50%
60%
70%
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
mea
n A
vera
ge P
reci
sio
n (
mA
P)
year
~1 year
~5 years
Region-based Convolutional Networks (R-CNNs)
[R-CNN. Girshick et al. CVPR 2014]
![Page 37: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/37.jpg)
Standard Neural Networks
𝒙 = 𝑥1, … , 𝑥784𝑇 𝑧𝑗 = 𝑔(𝒘𝑗
𝑇𝒙) 𝑔 𝑡 =1
1 + 𝑒−𝑡
“Fully connected”
![Page 38: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/38.jpg)
From NNs to Convolutional NNs
• Local connectivity
• Shared (“tied”) weights
• Multiple feature maps
• Pooling
![Page 39: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/39.jpg)
Just-in-Time Information• What is a convolution?
• In signal processing, a correlation is an operation that multiplies a small mask times a small piece of the image. These are examples of such masks.
• The strict definition of convolution flips the mask.
• But in computer vision, we call everything convolution.
![Page 40: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/40.jpg)
Convolutional NNs
• Local connectivity
• Each green unit is only connected to (3)neighboring blue units
compare
![Page 41: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/41.jpg)
Convolutional NNs
• Shared (“tied”) weights
• All green units share the same parameters 𝒘
• Each green unit computes the same function,but with a different input window
𝑤1
𝑤2
𝑤3
𝑤1
𝑤2
𝑤3
![Page 42: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/42.jpg)
Convolutional NNs
• Convolution with 1-D filter: [𝑤3, 𝑤2, 𝑤1]
• All green units share the same parameters 𝒘
• Each green unit computes the same function,but with a different input window
𝑤1
𝑤2
𝑤3
![Page 43: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/43.jpg)
Convolutional NNs
• Convolution with 1-D filter: [𝑤3, 𝑤2, 𝑤1]
• All green units share the same parameters 𝒘
• Each green unit computes the same function,but with a different input window
𝑤1
𝑤2
𝑤3
![Page 44: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/44.jpg)
Convolutional NNs
• Convolution with 1-D filter: [𝑤3, 𝑤2, 𝑤1]
• All green units share the same parameters 𝒘
• Each green unit computes the same function,but with a different input window
𝑤1
𝑤2
𝑤3
![Page 45: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/45.jpg)
Convolutional NNs
• Convolution with 1-D filter: [𝑤3, 𝑤2, 𝑤1]
• All green units share the same parameters 𝒘
• Each green unit computes the same function,but with a different input window
𝑤1
𝑤2
𝑤3
![Page 46: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/46.jpg)
Convolutional NNs
• Convolution with 1-D filter: [𝑤3, 𝑤2, 𝑤1]
• All green units share the same parameters 𝒘
• Each green unit computes the same function,but with a different input window𝑤1
𝑤2
𝑤3
![Page 47: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/47.jpg)
Convolutional NNs
• Multiple feature maps
• All orange units compute the same functionbut with a different input windows
• Orange and green units compute different functions
𝑤1
𝑤2
𝑤3
𝑤′1𝑤′2𝑤′3
Feature map 1(array of greenunits)
Feature map 2(array of orangeunits)
![Page 48: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/48.jpg)
Convolutional NNs
• Pooling (max, average)
1
4
0
3
4
3
• Pooling area: 2 units
• Pooling stride: 2 units
• Subsamples feature maps
![Page 49: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/49.jpg)
Image
Pooling
Convolution
2D input
![Page 50: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/50.jpg)
The key to SVMs
• It’s all about the features
Histograms of Oriented Gradients for Human Detection, Dalal and Triggs, CVPR 2005
SVM weights(+) (-)
HOG features
![Page 51: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/51.jpg)
Core idea of “deep learning”
• Input: the “raw” signal (image, waveform, …)
• Features: hierarchy of features is learned from the raw input
![Page 52: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/52.jpg)
• If SVMs killed neural nets, how did they come back (in computer vision)?
![Page 53: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/53.jpg)
What’s new since the 1980s?
• More layers• LeNet-3 and LeNet-5 had 3 and 5 learnable layers
• Current models have 8 – 20+
• “ReLU” non-linearities (Rectified Linear Unit)• 𝑔 𝑥 = max 0, 𝑥
• Gradient doesn’t vanish
• “Dropout” regularization
• Fast GPU implementations
• More data
𝑥
𝑔(𝑥)
![Page 54: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/54.jpg)
What else? Object Proposals• Sliding window based object detection
• Object proposals• Fast execution
• High recall with low # of candidate boxes
ImageFeature
ExtractionClassificaiton
Iterate over window size, aspect ratio, and location
ImageFeature
ExtractionClassificaiton
Object Proposal
![Page 55: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/55.jpg)
The number of contours wholly enclosed by a bounding box is indicative of the likelihood of the box containing an object.
![Page 56: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/56.jpg)
Ross’s Own System: Region CNNs
![Page 57: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/57.jpg)
Competitive Results
![Page 58: Object detection, deep learning, and R-CNNs · Object detection, deep learning, and R-CNNs Partly from Ross Girshick Microsoft Research Now at Facebook. ... Slides by Pete Barnum](https://reader034.vdocuments.net/reader034/viewer/2022042322/5f0c554c7e708231d434e166/html5/thumbnails/58.jpg)
Top Regions for Six Object Classes