We show that object detectors can hallucinate and detect missing objects; potentially even accurately localized at their expected, but non-existing, position. This is particularly problematic for applications that rely on visual part verification: detecting if an object part is present or absent. We show how popular object detectors hallucinate objects in a visual part verification task and introduce the first visual part verification dataset: DelftBikes 1, which has 10,000 bike photographs, with 22 densely annotated parts per image, where some parts may be missing. We explicitly annotated an extra object state label for each part to reflect if a part is missing or intact. We propose to evaluate visual part verification by relying on recall and compare popular object detectors on DelftBikes.
|Title of host publication||2021 IEEE International Conference on Image Processing (ICIP)|
|Subtitle of host publication||Proceedings|
|Place of Publication||Piscataway|
|Number of pages||5|
|Publication status||Published - 2021|
|Event||2021 IEEE International Conference on Image Processing (ICIP) - Virtual at Anchorage, United States|
Duration: 19 Sep 2021 → 22 Sep 2021
|Conference||2021 IEEE International Conference on Image Processing (ICIP)|
|City||Virtual at Anchorage|
|Period||19/09/21 → 22/09/21|
- Visual part verification
- object detection
FingerprintDive into the research topics of 'Hallucination In Object Detection: A Study In Visual Part VERIFICATION'. Together they form a unique fingerprint.
DelftBikes, data underlying the publication: Hallucination In Object Detection-A Study In Visual Part Verification
van Gemert, J. C. (Creator), TU Delft - 4TU.ResearchData, 1 Jul 2021