Supplementary Material for “Semi-supervised Learning for Large Scale Image Cosegmentation” Zhengxiang Wang Rujie Liu Fujitsu Research & Development Center Co., Ltd, Beijing, China {wangzhengxiang,rjliu}@cn.fujitsu.com

Abstract This supplementary material provides the detailed derivation of the three terms in the energy function to the binary quadratic programming (QP) problem, as well as its decomposition to sub-problems. The plots of the cosegmentation accuracy affected by the different values of parameters are also attached, showing that the proposed method is not sensitive to the choice of parameters within a range.

1. Detailed Equation Derivation We firstly show how the inter-image distance, intra-image distance and balance term of the energy function (Equation M1, M2 and M5 respectively, M stands for the equation label from the main manuscript, to distinguish the equation label in this supplementary material) are converted to form the binary QP problem (Equation M8, M13 and M16 respectively).

1.1. Inter-image distance The inter-image distance is converted from Equation M1 to Equation M8 by direct expansion: Einter

=

Nu X Ns X

k Hi · yi − Hjtr · yjtr k2 +

=

k Hi · yi − Hj · yj k2

i=1 j=i+1

i=1 j=1 Nu X Ns X

Nu X Nu X

[yiT · HiT · Hi · yi − 2 · yiT · HiT · Hjtr · yjtr + (yjtr )T · (Hjtr )T · Hjtr · yjtr ]

i=1 j=1

+

Nu Nu X X

(yiT · HiT · Hi · yi − 2 · yiT · HiT · Hj · yj + yjT · HjT · Hj · yj )

i=1 j=i+1

=

Nu X

yiT · (Ns · HiT · Hi ) · yi +

+

yiT · [(Nu − 1) · HiT · Hi ] · yi +

=

=

i=1

Hjtr · yjtr ) + Nu ·

Ns X

j=1 Nu X

Nu X

yiT · [(Ns + Nu − 1) · HiT · Hi ] · yi +

i=1 Nu X

Ns X

(yitr )T · (Hitr )T · Hitr · yitr

i=1

yiT · (−2HiT · Hj ) · yj

i=1 j=i+1

i=1 Nu X

yiT · (−2HiT ·

i=1

i=1 Nu X

Nu X

Nu X

yiT · Vi + C +

i=1

yiT · Miiinter · yi +

Nu X Nu X

inter yiT · Mij · yj +

i=1 j=i+1

Nu X Nu X

inter yiT · Mij · yj

i=1 j=i+1 Nu X

yiT · Vi + C

i=1

inter where Miiinter , Mij , Vi and C are defined in Equation M9, M10, M11 and M12 respectively.

1

(1)

1.2. Intra-image distance The intra-image distance is converted from Equation M2 to Equation M13 according to the fact that yi is a binary vector, we have: |yi (j) − yi (k)| = [yi (j) − yi (k)]2 (2) Therefore Equation M2 can be reformulated as: Eintra

=

si Nu X X

Wi (j, k) · [yi (j) − yi (k)]2

i=1 j=1,k=1

=

si Nu X X X

[Wi (j, k) · yi (j)2 + Wi (j, k) · yi (k)2 − 2Wi (j, k) · yi (j) · yi (k)]

i=1 j=1 k∈N (j)

=

Nu X si si X si X X X { yi (j)2 · [ Wi (j, k) + Wi (k, j)] − yi (j) · [Wi (j, k) + Wi (k, j)] · yi (k)} i=1 j=1

(3)

j=1 k=1

k∈N (j)

We can reformulate the above equation in the matrix form by constructing a matrix Miintra with the definition provided in Equation M14 and M15, so that the above equation can be converted to Equation M13: Eintra

=

Nu X si si X si X X X { yi (j)2 · [ Wi (j, k) + Wi (k, j)] − yi (j) · [Wi (j, k) + Wi (k, j)] · yi (k)} i=1 j=1

=

Nu X

j=1 k=1

k∈N (j)

yiT · Miintra · yi

(4)

i=1

1.3. Balance term The balance term is converted from Equation M5 to Equation M16 through Taylor expansion, we firstly rewrite Equation M5 as: Nu Nu X X Ebal = [Pif log Pif + (1 − Pif ) log(1 − Pif )] = f (Pif ) (5) i=1

i=1

where f (x) = x log x + (1 − x) log(1 − x)

(6)

Since x = Pif ∈ [0, 1] is continually derivative, we can use Taylor expansion to approximate f (x) at x = 21 : f 00 ( 12 ) 1 1 1 1 (x − )2 f (x) ≈ f ( ) + f 0 ( )(x − ) + 2 2 2 2 2 1 = −1 + 2(x − )2 2 1 2 = 2x − 2x − 2

(7)

Therefore Ebal

=

Nu X

f (Pif )

i=1

= =

Nu X

1 [2(Pif )2 − 2Pif − ] 2 i=1

Nu X i=1

(2

yiT · ei yiT · ei · eTi · yi 1 − 2 − ) 2 si si 2 2

(8)

1.4. The whole energy function By summing all these three terms, the whole energy function E of Equation M17 can be derived by: E

= Einter + λ1 · Eintra + λ2 · Ebal Nu Nu Nu X Nu X X X inter = yiT · Miiinter · yi + yiT · Mij · yj + yiT · Vi + C i=1

+ λ1

i=1 j=i+1

Nu X

yiT · Miintra · yi + λ2

i=1

=

Nu X

Nu X

(9)

i=1

(2

yiT

·

i=1

yiT · (Miiinter + λ1 Miintra + 2λ2

i=1

ei · eTi s2i

· yi

−2

yiT · ei 1 − ) si 2

Nu X Nu Nu X X 1 ei · eTi ei T inter ) · y + y · M · y + yiT · (Vi − 2λ2 ) + C − i j i ij 2 si si 2 i=1 j=i+1 i=1

Removing constant terms we can get Equation M17.

1.5. Decomposition to sub-problems The decomposition of the whole energy function (Equation M17) to sub-problems (Equation M20 is straightforward: Ei

=

Nu X

yiT · (Miiinter + λ1 Miintra + λ2

i=1

=

Nu X

Nu X

yiT · Mi0 · yi + yiT ·

i=1

Nu X Nu Nu X X ei · eTi ei T inter ) · y + y · M · y + yiT · (Vi − λ2 ) i j i ij 2 si si i=1 j=i+1 i=1

inter Mij · yj +

= yiT · Mi0 · yi + yiT ·

inter Mij · yj + yiT · (Vi − λ2

j=1,j6=i

+

Nu X

[yjT

·

Mj0

· yj +

yjT

Nu X

·

= yiT · Mi0 · yi + yiT · (

Nu X

j=1,j6=i

=

·

Mi0

· yi +

yiT

Nu X i=1

yiT · (Vi − λ2

ei ) si

ei ) si

inter Mjk · yk + yjT · (Vj − λ2

inter Mij · yj + Vi − λ2

inter Mjk · yk +

k=j+1,k6=i

k=j+1,k6=i

j=1,j6=i

yiT

Nu X

yjT ·

j=1,j6=i

j=1,j6=i Nu X

Nu X

ej )] sj

ei ) + Ci0 si

· Vi0 + Ci0

(10)

where Mi0 and Vi0 are defined in Equation M21 and M22 respectively and Ci0 is the constant term for sub-problem Ei .

2. Experiment We also show in Figure 1 the plots of the cosegmentation accuracy affected by the choice of parameters (λ1 and λ2 ), using “Baseball” class in iCoseg dataset as example. We can see that the accuracy would be stable when λ1 is within a range of its best choice. For λ2 , increasing its value will increase the accuracy until reaching the best result, and further increasing it after the best choice would be slightly lower than the best accuracy but still stable. This results validate that the proposed method is not sensitive to the choice of parameters within a range.

3

Figure 1. Accuracy under different parameter values for “Baseball” class in iCoseg dataset. Top: λ1 ; Bottom: λ2 .

4

Supplementary Material for

Fujitsu Research & Development Center Co., Ltd, Beijing, China. {wangzhengxiang,rjliu}@cn.fujitsu.com. Abstract. This supplementary material provides the ...

250KB Sizes 3 Downloads 267 Views

Recommend Documents

Supplementary Material
Jan Heufer ∗. *TU Dortmund University, Department of Economics and Social Science, 44221 Dortmund,. Germany. .... 3 More Details on the Data Analysis.

Supplementary Material for
Aug 3, 2016 - alternatives are “autocracy and risk-sharing” and “democracy and investment.” In other words, the .... whether seizing power. A coup is .... Our main sources are Galasso (1976), Stearns (2001), and Ortu (2005). References.

Supplementary Material for
Sep 25, 2015 - Using archived and field-collected worker bumble bees, we tested for temporal changes in the tongue ... (Niwot Ridge Long Term Ecological Research Site, 40°3.567'N, 105°37.000'W), and by. P. A. Byron (17) ... these transects did not

Supplementary Material
gaze fixation point is detected by an eye tracker. ∗indicates equal contribution. (a) Random ... 1) confirms quanti- tatively our intuition, that the location of the hands are in- deed important for learning about hand-object interactions. Based on

Supplementary Material - Arkivoc
General Papers. ARKIVOC 2015 (vii) S1-S13. Page S3. ©ARKAT-USA, Inc. δ /ppm. 1. H Assignment. 8.60 (brs, 1H). 5.35 (t, J = 7.3 Hz, 1H). 5.08 (t, J = 6.1 Hz, ...

Supplementary Material
By the definition of conjunctive pattern, since S does not satisfy pc, then S ... Then there exists at least one instance of p o in S: {I1,I2,...,Im},. I1 ∈ I(p c. 1 ,S),...,Im ∈ I(p c m,S), such that ∀t1 ∈. I1,..., ∀tm ∈ Im,t1 < t2 < ...

Supplementary Material
and Business Cycles Conference, the Bank of Canada 2014 Fellowship ..... CGH exhibited an upward trend in the frequency of sales that could explain why they find ...... 12Fox and Sayed (2016) use the IRI scanner dataset to document the ...

Supplementary Material
The data are provided a for 10 subsectors, which I aggregate up to three sectors as follows. ... CAN Canada ...... Structural change in an open economy. Journal ...

Supplementary Material
SVK Slovakia. CYP Cyprus. ITA. Italy. SVN Slovenia. CZE Czech Republic JPN Japan. SWE Sweden. DEU Germany. KOR South Korea. TUR Turkey. DNK Denmark. LTU Lithuania. TWN Taiwan. ESP Spain. LUX Luxembourg. USA United States. EST Estonia. LVA Latvia. RoW

Supplementary Material for ``Observations on ...
Nov 19, 2017 - C.2 Steady State in a Perturbed Environment. In this subsection we formally adapt the definitions of a consistent signal profile and of a steady state to perturbed environments. Let f((1−ϵ)·σ+ϵ·λ) : OS → OS be the mapping bet

SUPPLEMENTARY MATERIAL FOR “WEAK MONOTONICITY ...
This representation is convenient for domains with complete orders. 1 .... check dominant-strategy implementability of many classical social choice rules. In.

SUPPLEMENTARY MATERIAL FOR “WEAK MONOTONICITY ...
This representation is convenient for domains with complete orders. 1 ... v = (0,v2,0), v2 > 0, would want to deviate and misreport their type so as to get 3.

Efficient Repeated Implementation: Supplementary Material
strategy bi except that at (ht,θt) it reports z + 1. Note from the definition of mechanism g∗ and the transition rules of R∗ that such a deviation at (ht,θt) does not ...

Supplementary Online Material
Branstetter, M.G. (2009) The ant genus Stenamma Westwood (Hymenoptera: Formicidae) redefined, with a description of a new genus Propodilobus. Zootaxa,.

Electronic supplementary material
Jun 22, 2009 - ... of two Hill functions (a-b) with sufficiently distant midpoints is equivalent to the Hill function with the largest midpoint (c). Namely: Hill(x, K 1) · Hill(x, K 2) ≈ Hill(x, K 2) where K 1

Supplementary Material - HEC Montréal
... the ONS website: http://www.ons.gov.uk/ons/datasets-and-tables/index.html. .... sitivity of sales to the unemployment rate for independent stores and chains.

Supplementary Material for ”Production-Based Measures of Risk for ...
Measures of Risk for Asset Pricing”. Frederico Belo. ∗. November 3, 2009. Appendix A Producer's Maximization Problem. Define the vector of state variables as xit-1 = (Kit-1,ϵit-1,Pit-1,Zit-1), where Kit-1 is the current period stock of capital,

Supplementary Material for ”Production-Based Measures of Risk for ...
Nov 3, 2009 - [4] Campbell, John Y., and Robert J. Shiller, 1988, Stock prices, earnings, and expected dividends, Journal of Finance 43,661 − 676. [5] Campbell, J., 2003, Consumption-Based Asset Pricing, in George Constantinides, Milton. Harris, an

Supplementary material for “Complementary inputs and ...
Aug 2, 2017 - Figure S1: The network structures produced in the proof of Theorem 3. Undirected ...... without loss of generality that has full support on .

Supplementary Material for Adaptive Relaxed ... - CVF Open Access
1Department of Computer Science, University of Maryland, College Park, MD ... 3Department of Mathematics, Hong Kong Baptist University, Kowloon Tong, ...

Supplementary material for “Complementary inputs and ...
Aug 2, 2017 - statement of Theorem 2, and then complete the proof. ...... Jones, 1984), spaces of square-integrable functions (Harrison and Kreps, 1979; ...

Supplementary Material Methods Tables
αF was added (25nM final) to exponentially growing MATa bar1Δ cells in YPD (OD600 ≈ 0.3) and the culture incubated for two hours at. 30oC. >95% G1 arrest, as indicated by cells with small buds (schmoo formation), was confirmed by microscopy. Cell

Supplementary Material: Adaptive Consensus ADMM ...
inequalities together. We conclude. (Bvk+1 − Bvk)T (λk+1 − λk) ≥ 0. (S3). 1.2. Proof of ..... matrix Di ∈ Rni×d with ni samples and d features using a standard ...

Price Selection – Supplementary Material
†Email: Departamento de Economia, Pontifıcia Universidade Católica do Rio de .... and-tables/index.html, the Statistics Canada's Consumer Price Research ...