Connect:  SPS Facebook Page  SPS Twitter  SPS LinkedIn  SPS YouTube Channel  SPS Google+ Page     Share: Share

Information Forensics and Security Technical Committee

ICASSP 2014 IFS-TC Meeting Minutes

Thursday May 8, 2014, 13:00-15:00 (GMT+1)
At ICASSP’13, Florence, Italy
Minutes prepared by G. Doërr and M. Wu

IFS-TC voting members: Gwenaël Doërr, Patrick Bas, Marco Tagliasacchi, Hitoshi Kiya, Jiwu Huang, Anthony Ho, Yan Sun, Samson Cheung, Tanya Ignatenko, Dinei Florencio, Matthias Kirchner, Oscar Au, Peter Hong,
Skype: Min Wu
Guests: Patrizio Campisi, Teddy Furon, Nasir Memon, Jean-Luc Dugelay, Mauro Barni, Slava Voloshynovskiy, Xiangui Kang

1. G. Doërr called for the meeting at 13:00

2. Roundtable for introductions

3. Updates from the Technical Direction Board

G. Doërr reports on a number of items that had been discussed during the Technical Direction Board (TDB) meeting a few days earlier. And the TC meeting participants discussed some of these issues as summarized below.

  • “The Power of a Name” discussion. The Society is currently contemplating the possibility of changing its name for a larger impact to general public, who is hardly aware of what ‘signal processing’ refers to. The initiative received mixed feedback from the different Technical Committees (TC) and the Society has therefore opened up a blog on the SPS website to collect feedback from a wider audience. G. Doërr encouraged the members of the IFS TC to go there on post their own view about the matter.
  • Internet of Things and Big Data. The Society created a new type of administrative entity referred to as Special Interest Groups (SIG). A SIG should be viewed as a light weight TC with a limited life-span (typically 3 years) to quickly address emerging technical areas. Successful SIGs will have the opportunity to convert into a TC on a case-by-case basis. The first two SPS SIGs will deal with the Internet of Things and Big Data.
  • FluidSurveys. The TDB subscribed to an online tool to conduct elections and polls. Each TC will have the opportunity to obtain an account to run various ballots and elections. G. Doërr encouraged in particular the Nominations and Elections Subcommittee to evaluate the tool to assess whether it could streamline the election process of new TC members.
  • SPS acceptance rate policy. After observing a few discrepancies over the last few years, the chair of the Conference Board announced the new policy relating to the acceptance rate of SPS flagship conferences: ICASSP and ICIP. The target is to have 45% of the submitted papers published with a flexibility of +/-3%. Historically, ICASSP is on the higher end and ICIP on the lower end.
  • EDAS conference management. The chair of the TDB announced that SPS is currently collecting feedback regarding the use of EDAS for ICASSP and ICIP. After a few hiccups, it looks like things are mostly under control, and there is room for improvement with respect to the user interface. Several TC requested the addition of a feature to see the reviews of the other reviewers, which may be an option to activate in the configuration of the system. M. Barni mentioned that there is significantly less support to a conference team for adopting EDAS (compared to some other full-service conference management/support options), and therefore, in terms of conference organization, it is recommended to have a dedicated chair to take care of the system if EDAS is chosen for paper and registration handling.
  • EDAS conference management.The information related to seeing reviews of others triggered a number of conflicting comments from the members of the IFS TC. Some people do not see the interest of such visibility and will actually not look at it. Some people highlighted the risk of biasing the reviewing process and stressed that if such visibility shall be provided in the future, it should only occur once the final decision has been made. Representatives of the MMSP TC reported on their own experience with a rebuttal phase for ICME and mentioned that it notably lengthen the reviewing process for a marginal impact on the final decision. G. Doërr will compile all the comments from the IFS TC and send a feedback email to the TDB.

4. Conference Activities

G. Doërr reports that 56 papers out of 122 had been accepted for publication in the IFS track at ICASSP 2014. This is a record number of submissions and may be due to the nice and central location (Florence, Italy). Still, a number of security-related papers were submitted in various other sessions, e.g. papers dealing with physical security. Relevant IFS TC members are encouraged to connect with this community to make the community aware of the IFS activities (e.g. WIFS) and encourage their participation.

G. Doërr reports that ICIP 2014 attracted nearly 300 submissions in the IFS track. This is also a record number of submissions and may be due to another nice location (Paris, France). The reviewing process has been delayed for various reasons, but the final decision notifications should be sent out soon.

J. Huang reports on WIFS 2013, previously held in December in China. The surplus reaches 11%, below the usual SPS target of 20%. Jiwu is experiencing difficulties to officially close the book of the workshop due to the departure of a person in the IEEE SPS staff. This is simply a matter of identifying the correct contact and will be sorted out shortly.

Y. Sun reports on the progress of WIFS 2014. After several delays, it is now official: GlobalSIP will host WIFS in Atlanta in 2014. The objective is to leverage on the format of the symposium to lower the organizational burden on our side. For information, the open call attracted two alternate security-related proposals, one on the security of the Internet of Things (IoT) and the other one on Active Authentication (AA). In view of the gentleman agreement with GlobalSIP that there should be no security-related competing event, we proposed to incorporate the two proposals as dedicated tracks within WIFS. The IoT proposal led by N. Memon (past IFS member) accepted while AA declined. Next, Yan showed the floor plan of the conference center and indicated the rooms which have been allocated to WIFS, essentially the keynote area that is unused besides GlobalSIP plenary talks. The organization team also intends to have short tutorial/overview talks over lunch break due to logistic constraints that preclude having a full day dedicated to tutorials as in previous editions. Yan then elaborates on the fee structure. WIFS authors will have to pay for the general GlobalSIP registration and a WIFS premium ticket. The premium ticket will provide access to WIFS plenary talks, discussion panels, and social event. It will be sold as a separate item to all attendees at registration. Regarding the social event, there are currently two options: a cost-effective one in the club room of the conference center and another one at the world of Coca-Cola. On this occasion, G. Doërr reminds that all TC members should help the organization team to make this edition a success by (i) publicizing the call for papers in relevant research communities in order to attract submissions and (ii) forwarding contacts of companies that are likely to financially support the event with a grant. Following up a comment, G. Doërr will also investigate how to guarantee that WIFS proceedings appear separately in IEEExplore.

G. Doërr asks P. Comesaña and his conf/workshop subcommittee to prepare and launch the call for proposals for WIFS 2015. P. Campisi expresses a potential interest to organize it in Roma, Italy. N. Memon mentions that he could also have an option in Abu Dhabi, United Arab Emirates.

5. Awards/Elections Activities

G. Doërr reminds on behalf on M. Wu that there is an on-going call for nominating Distinguished Lecturers. The IFS TC has historically quite successful with this award. A number of renowned individuals have been approached but, due to other commitments in the year to come, they declined our invitations so far. As a result, we are still looking for strong candidates. The voting procedure will follow-up shortly.

G. Doërr then announces that the call for nominations for SPS major awards will be issued soon: (SPM/SPL) best paper award, best young paper award, overview paper award, sustained impact paper award, etc. Last year, the IFS TC (and TIFS EB)’s endorsed nominations were unsuccessful in the ultimate award selection on the SPS level due to increased competitions. The TDB is currently working on harmonizing nomination practices across TCs but the outcome of this consultation may come in to late for this year. In this case, the IFS TC will proceed with the same rules as last year e.g. endorsement/co-nomination thresholds, management of conflict of interests, etc. In view of the increasing importance given to citation metrics, it is crucial to move forwards nominations that have a chance to compete at the Society level. As raised by T. Furon, this is particularly challenging for our small-size community that inherently put the citation record for IFS publications in a disadvantage position.

Finally, G. Doërr reminds that the TC will renew one third of the TC members as usual and will also elect its Vice-Chair this year. The nomination period is expected to run during the second half of the summer with a ballot in Fall. All TC members are encouraged to identify potential motivated candidates that would be willing to participate in the IFS TC. Based on the experience of past years, Gwenaël reminds that participating to various IFS activities (contest, reviews, conferences, etc) prior to the ballot is definitely a plus to be known by the current members and thereby attract more votes.

6. Contest Activities

G. Doërr reports on behalf of A. Rocha that the organizers of the first IFS challenge run in 2013 are looking into publishing a column in the Signal Processing Magazine to make the result of the contest known to a wider audience. The website hosting the contest will also be maintained for interested researchers who would like to benchmark their latest forensic algorithms. Afterward, Gwenaël probes whether there would be interest in organizing a second contest. In view of the efforts dedicated to the first contest, the TC members agree not to organize a second edition immediately. N. Memon mentions that organizing such contests from scratch is highly time-consuming and requires a lot of dedication. He mentions that NYU set up a platform to host capture-the-flag contests. The IFS-TC may consider leveraging on this platform to facilitate the organization of future IFS contests.

G. Doërr then highlights the importance of participating to activities at the Society level which provides greater visibility. The SP cup competition could be one option and Gwenaël mentioned that M. Wu is considering putting in a proposal related to IFS. Min then presents a rough plan of her tentative proposal one electric power signatures (ENF), which is organized in two main tasks: building an ENF measurement apparatus and exploiting the obtained recordings to answer forensic questions e.g. the place and time of the acquisition. The objective would be to tap in the basic circuit/electronics, data acquisition/sampling, and signal processing knowledge of undergrad students at junior and beginning senior level. She plans to investigate with potential partners in different parts of the world if there is a possibility to team up to run this contest and will get back to the TC in case it is feasible. G. Doërr indicates that the timeline is extremely tight and that the proposal may be postponed to next year.

7. Website Activities

Now that the IFS website has been equipped with social media capabilities, it is important to provide content to keep it alive. G. Doërr encourages all TC members to liaise with the website subcommittee and to forward relevant information (special issues, call for papers, open positions, news announcements) so that it could be publicized. He also encourages them to proactively look for potential individual that could be enrolled in the communities on the different social platforms, both from the academic world and the industrial arena.

8. Community Building

While the IFS TC has one of the highest affiliate members per published paper in ICASSP ratio in SPS, there is definitely room for improvement. There is momentum building in the Society to go beyond scientific contributions only. In particular, they are looking for initiatives that would provide meaningful services to signal processing practitioners in the industry and student members. It is critical to sustain the growth of the Society, and G. Doërr highlights that the TC should also embrace this effort at the IFS level. The IFS community indeed remains relatively small and there is a growing concern that we might not reach the critical mass to be self-sustainable. Each TC member should therefore look into means to build the community. One idea would be to further enlarge the scope of the TC. Another idea is participate in high visibility initiatives such as SigView and seasonal schools. In particular, keynotes given at WIFS and lectures given by IFS-endorsed Distinguished Lecturers should definitely be recorded to raise awareness and stimulate interest.

9. Publication News

TIFS is doing well in terms of submissions, and its citation influence is improving from last year. Jay Kuo will retire at the end of the year. The next Editor in Chief shall be announced shortly.

The deployment of SigPort is under way. M. Wu and Y. Sun have been involved in this SPS initiative. Y. Sun presents a short video clip prepared by M. Wu illustrating the basic functionality of a pilot system.

10. Next meeting

The next TC meeting will be held in Atlanta, Georgia, USA, during WIFS 2014 (December 3-5, 2014). The meeting was adjourned at 15:00 without objection.