Many ambiguous objects are difficult to be recognized or detected in real life scenarios. Recently, there exist new trends of detecting such challenging objects, for example, camouflaged, transparent, or in-mirror objects in computer vision that attract lots of researchers’ attention. In fact, this problem is very challenging yet interesting in the research field due to lighting, occlusion, size variance, and ambiguity. Another notable problem lies on the rare data and small datasets. In practice, this research can be applied into different applications, i.e., wild-life preservation, camouflage material, search and rescue mission, or even medical image analysis.

The main objective of this workshop is to advance multimedia research in analyzing ambiguous objects, including camouflaged, transparent, in-mirror, and amodal objects, etc. The ability to autonomously identify such objects plays a crucial role in various domains, such as search-and-rescue operations, wildlife discovery and conservation, and medical diagnostics (e.g., polyp detection and segmentation, COVID-19 identification from lung X-rays). In multimedia forensics, advanced techniques are essential for detecting and segmenting manipulated images and videos, combating deep fake content, and ensuring digital media integrity. Moreover, advancements in ambiguous object recognition can enhance augmented reality (AR) and virtual reality (VR) experiences, improve scene understanding in video analysis, and refine content moderation systems for social media platforms.

We welcome people to submit papers about ambiguous object analysis to AOCV workshop. Accepted papers will be presented in our workshop and will be published in the ACM MM workshop proceeding.

We accept short papers (non-archived) which are up to 4 pages in ACM MM format, excluding references; and long papers (archived) which are up to 8 pages in ACM MM format, excluding references. Submission policies adhere to the ACM MM 2025 submission policies.

The topics in this workshop will include but are not limited to:

• Ambiguous object recognition
• Ambiguous object detection
• Ambiguous object segmentation
• Ambiguous object tracking
• Ambiguous object Re-ID
• Ambiguous object generation
• 3D reconstruction
• Behavior analysis
• Uncertainty analysis
• Large-scale datasets
• Data augmentation
• Applications

• Paper submission deadline: Jun. 13, 2025 Jul. 11, 2025 (Extended)
• Acceptance notification: Aug. 01, 2025
• Camera-ready deadline: Aug. 10, 2025
• Workshop date: Otc. 27-28, 2025