Backend
SurgCBVR: A Multi-Domain Benchmark for Content-Based Surgical Video-to-Video Retrieval
arXiv eess.IV0 views
arXiv:2611.23347v1 Announce Type: new Abstract: Hospitals record thousands of hours of surgical video annually, yet these archives remain difficult to search by content. Surgeons cannot readily retrieve similar operations, complications, or workflow events. Existing surgical computer vision primarily produces labels rather than retrieving relevant videos. We introduce SurgCBVR, the first benchmark for content-based video-to-video retrieval in surgery. It integrates eight public datasets across four domains-ophthalmic, laparoscopic, robotic-assisted, and colonoscopy-covering six procedure types, 679 cases, and 76,489 clips (106 hours). SurgCBVR comprises two tiers: Scene, retrieving the same operative state across cases and sites, and Workflow, retrieving and localizing phase transitions and deviations. Scene includes 74,765 clips spanning 40 classes, three temporal granularities, and three playback speeds; Workflow contains 1,724 transition clips. Fixed clip lengths and frame counts, together with phase-classifier oracle and duration-only baselines, help assess label and timing shortcuts. We evaluate six generic and surgical foundation models, two video retrieval methods, and contrastive heads trained on SurgCBVR, using within-dataset, cross-dataset, and pooled-gallery retrieval restricted to the same procedure type. Frozen representations barely outperform a timing-only baseline, transition retrieval approaches chance, and trained heads lose up to half their accuracy across datasets. These findings expose substantial limitations in current video representations and establish SurgCBVR as a benchmark for advancing surgery-aware retrieval.
Source:https://arxiv.org/abs/2611.23347PreviousBeyond Accuracy: Benchmarking Uncertainty Quantification in Retinal Disease ScreeningNextSWIMSS: An Open Multi-Center SwinUNETR Model for Multiple Sclerosis White Matter Lesion Segmentation
Related posts
Why Did Google Swap Version Numbers for the Periodic Table? (Gemini 4: Hydrogen, Neon, and Carbon)
0dev.to AI
I describe an n8n workflow in plain English and Claude Code builds it: the setup and the 4 rules that stop it breaking things
0dev.to AI
An AI sales assistant on Telegram that knows when to hand off
0dev.to AI
Build a Crash-Safe Scheduled Publisher with Leases and Reconciliation
0dev.to WebDev
OpenRouter Routed Some Chatbot Prompts to a Hacked Server for 18 Days. Were Your Passwords in Them?
0dev.to Security
I Built a Java Agent That Injects Tenant Filters Into Every SQL Query Automatically
0dev.to Security