Subject Detection and Manipulation!Eric Chang, Austin Lee and Samuel Yang!

Department of Electrical Engineering, Stanford University

Motivation Image Processing Pipeline

Evaluation Metric

Results

Many interesting artistic manipulations can be done on images and video if the subject can be reliably detected: •  background blurring •  automatic refocusing, exposure, white balance •  object image search Here we implement and extend existing methods for detecting salient objects. We then use the result to demonstrate an interesting artistic manipulation, whereby we desaturate the background and generate a simulated stereo view of the subject, viewable as an animated GIF.

contrast color spatial distribution

center surround average

contrast color spatial distribution

center surround average

contrast color spatial distribution

center surround averagecolor1 contrast1 center surround1

color2

1 Zheng, N., & Tang, X. (2007). Learning to Detect A Salient Object, 1–8. IEEE. 2 derived masks eveloped for the project 3 Goferman, S., Zelnik-Manor, L., & Tal, A. (2012). Context-aware saliency detection. IEEE PAMI.

content aware3

original image final segmentation map desaturation map final desaturated image

contrast2 content aware2

original image final segmentation map desaturation map final desaturated image

Saliency maps

segmentation input

We used a subset of 200 images out of the 20,000 in [1], in which each image has exactly one salient object, and for which a ground truth bounding box of the salient object is known. Using Jaccard similarity of the bounding boxes as our performance metric, we increased accuracy to 68% over 200 images by using the three derived saliency maps and our new segmentation scheme.

original saliency map

desaturation map desaturated image

poor examples

hist

ogra

m c

ount

output