Project Number 14: Android based Object Detection and Classification: Modeling a Child’s Learning of What Is Hot and Cold Should this project be considered for the Best Project Award? No. Overall, though the project idea is interesting and contains potentially novel findings, the project implementation is subpar and the final report was organized in a hard to follow manner (please see the comments below). Despite the draw backs and challenges, the proposed approach was not followed during the implementation. With that said, since it is only a one person project, it is more challenging to thoroughly consider all the draw backs and challenges and analyze all the collected data for this project. An attempt was made to create the final report as the IEEE paper publication standard; however, by doing so, it had hindered the overall report presentation. Should this project be considered for the top 3 project awards? No. In addition to the reasons provided to the first question, please see the answers below. On a scale of 1 to 10, how would you rate the overall organization/clarity of the project? 6.5: The project report is hard to follow, and most of its diagrams were hard to read, and many of the diagrams provided little or no value. The report was not proof read before submission and thus contained many spelling errors and grammar mistakes. The format between diagrams and text was done poorly in some parts of the report. Also some of the contents for its headings are incomplete or incorrect. Some of the information on the project was repeated several times and added no value. This project report, however, contains headings that helped to divide the contents into a more readable manner. More Reasons: • The Abstract section should provide relevant information about the project, defines the project scope approach, and gives a summary of the project results. However, the abstract provided here gives no indication of the project results and is ambiguous about the project approach. • There are many grammar errors and misspelled words, such as during the Abstract: the phrase “hot or code” should be changed to “hot or cold” • The Index Terms should include embodiment, computational perception, machine vision and along with some other. • The Introduction should provide the background information the reader will need to understand the project. It should also describe how the rest of the project will be presented. The Introduction section here does a poor job of giving the project background and scope and it failed to address the audience. “The project is targeted to students who have simple technology like their cell phone and micro-controller kits available to run experiments on”. First of all, this should have its own section: Intended Audience. This section should include who this project will benefit, who is interested in the results, and why this is important to them. • While it is good to have a Changes From Proposal section, many changes were mentioned elsewhere on the report. It is good to consolidate the information to avoid redundancy. • The Existing Research section contains some good information but the presentation/delivery was subpar. A better way of presenting this section is to divide the related works into a few different categories, such as related work on affordances, related work on embodiment, related work on self-organizing map, etc. • The Approach section contains good information but again, it needs to be checked for grammar errors and spelling mistakes before the final submission. The use case diagrams are good; they provide good information to the reader. However, figures like figure 10 contains too much information on one figure and it will be more effective if a section was zoomed and explained more. Figures like figure 10 should be included just once; after that, it does not provide any additional values. • The Experiments section was only half a page and can be expanded. In this section data collection methods can be elaborated and some of the changes from the proposal can be elaborated here as well. Some of the information from the Approach section should be moved into this section. • The results section contained good information as well; however, the data collection section from results should be in the Experiment section and the data analysis portion should be included in the results. More charts and graph is needed in this section to help the reader to understand the result. There should be charts that let the users know right away about the result of this project. Figures like figure 19 might contain some good information but it is hard to read and there are too many pictures inside for it to be effective. A different approach needs to be used here. • The Future Research/Lessons Learned section contains good information. To better help the reader to know what future research is covered here sub headings should be included, such as: Algorithm Implementation, Bias Addition. The same thing applies to Lessons Learned section as well. • The Conclusion section contained contents and did a good job wrapping up. • The References section needs to be formatted between with its alignment. When a word is spread out this far, it is really hard to read and unprofessional. • For Appendix A, Figure 36 needs to be moved up and aligned correctly with the rest of the figures. There should be more descriptive names than image_64.jpg. For example, image_64.jpg can be named to pink_marker2.jpg On a scale of 1 to 10, how would you rate the overall project idea? 8: The project idea is interesting, relevant to Developmental Robotics, and contains potential novelty. The novelty for this project is taking the temperature into a qualification method along with visual. However, this project did not really use temperature very effectively. All the objects used contained close to the same temperature, and thus reduced the point of this project. On a scale of 1 to 10, how would you rate the overall research contribution of the project idea, methodology and/or results? 8.5. Aside from the answers to the previous questions, the amount of work involved for this project is significant and yielded some good results. The data was collected and but not well explained during the final report. Unfortunately I was not assigned for the proposal review for this paper. I would have made some suggestions during the proposal and would love to see them implemented during the project implementation. Concerns, issues, questions, and proposals • Experiment details o It should be mentioned the operational temperature range of the temperature sensor and its accuracy. o The temperature range between objects needs to be varied more than just by 10 or 15 degrees Fahrenheit. It defeats the purpose of using the temperature as a measure to distinguish objects when all the objects classified are very close in value for temperature. Question: is a marker hotter or cooler than a pill box? Answer: it largely depends on the room temperature of the two objects. o There needs to be a set condition, such as all the objects were tested and sampled in the same room with the same environmental temperature. To test objects with varying temperatures, an assumption can be made that for some objects such as ice, that it will always be cold, and for objects such as a lighter (when used) will always be hot. o There should be results gathered for classification without considering temperature, results gathered for classification with only temperature to compare and contrast the effectiveness of the results. • Others o The words “learning” and “what’s” should be capitalized and since the project did not really empathize on the “Child’s Learning” portion, where the data was “feed” into the system, this title needs to be changed to reflect this.