| Definitely accept Weakly accept Borderline paper Weakly reject Definitely reject |
For the sake of the authors and the quality of the reviewing process, please explain your ratings in the space provided. Your rating should not consider whether or not the paper might become an oral or a poster. |
| Very confident Confident Not confident |
By accepting a paper for review, the reviewer confirms his competence in the required areas. He is thus expected to be quite confident in his/her conclusions. "Confident" is therefore the default option. Use the other options to stress that you are absolutely sure about your conclusions (you are an expert in the respective area) or that you feel some doubt (you are not an expert in the respective area, but you are able to distinguish good work form a bad work in the respective area). If you have serious doubts about your ability to assess the paper, please inform the PC chairs. |
| Very original Moderately original Minor originality Has been done before (implies reject, justify thoroughly) |
This is the standard notion of novelty. "Very original" papers open new directions and often become seminal papers. The "Has been done before" must be accompanied by relevant references. |
| Of broad interest Of sufficient interest Of limited interest Irrelevant or out of scope for ICCV (implies reject, justify thoroughly) |
Every researcher in CV should find interest in works "Of broad interest" for, e.g., a contribution in his field of interest, the technical quality of the work, or a surprising result. Works "Of sufficient interest" do not have to address everyone in the audience, but should have an impact in a certain area. Works "Of limited interest" should be considered for ICCV only if their novelty, clarity, and correctness is excellent. |
| Excellent reference to prior work References adequate References missing Does not cite relevant work (implies reject, justify thoroughly) |
A "Does not cite relevant" work strongly suggests reject. This option should be selected if the missing work is well known in the community and commonly cited, else we suggest giving the authors the benefit of the doubt by selecting "References missing". List the missing references. |
| Reads very well Is clear enough Difficult to read Unreadable (implies reject, justify thoroughly) |
Difficult to read: text is ambiguous at several points because explanations are missing or difficult to follow. |
| Definitely correct Probably correct, did not check completely Contains minor errors Has major problems (implies reject, justify thoroughly) |
The statement that a paper is "Technically correct" means that its conclusions are supported by flawless arguments. Proofs are correct, formulas are correct, there are no hidden assumptions, experiments are well designed and properly evaluated. |
| Sufficient experimental validation or a theoretical paper Limited but convincing Lacking in some respect Insufficient validation |
Different papers need different levels of experimental validation. A theoretical paper may need no experiments. A paper presenting a new idea might just need an experiment illustrating that there exists a situation where the idea applies. A paper presenting a new phenomenon or a performance evaluation paper may need a thorough experiments and their evaluation. |