Review Instructions

Review instructions

You should write a review that you would like to receive yourself. Your reviews should be helpful to the authors, even if the review recommends the rejection of the paper. Papers do not need to be focused on applying AI techniques to multimedia data. Our research community is rich and diverse, and papers addressing many aspects of multimedia analytics and retrieval are welcome. Even though all reviews are anonymous, we hope that you prepare excellent reviews, reviews that you would be proud to associate with your name. Confirm that the paper that you are reviewing falls into the topical scope of ICMR, as defined by the Call for Papers. Eventually, we rely on your judgment and the collective wisdom of your peers to decide if the paper aligns with multimedia topics.

  • Reviews should not just state, “It is well known that…”, but rather, they should include citations. Likewise, reviews should not just state, “Important references are missing…”, but rather, they should include examples.
  • Reviewers should list their own references only in very rare cases when these are indeed the most relevant references for the authors to refer to.
  • Reviews should not just state, “Authors should compare to the state of the art…”, but rather, they should cite specific work (i.e., peer-reviewed references) that they feel the authors should have considered and why.
  • Reviews should critique “the paper” and not the authors.
  • Reviews should try not to address the authors directly, esp. not as “you”. (A direct address can be interpreted as an affront by the reader).
During the review process, no attempt should be made to guess the identity of the authors.