DORI benchmark shows top vision-language models reach only 54.2% accuracy on coarse orientation tasks and 33% on granular judgments, with sharp drops on reference-frame shifts and compound rotations.
In: Computer Vision– ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2verdicts
UNVERDICTED 2representative citing papers
SSD MobileNet V1 shows lowest latency and energy but lowest accuracy while YOLOv8 Medium shows highest accuracy at higher cost; Jetson Orin Nano gives best overall balance and accuracy gaps widen with increasing scene complexity.
citing papers explorer
-
Seeing Isn't Orienting: A Cognitively Grounded Benchmark Reveals Systematic Orientation Failures in MLLMs
DORI benchmark shows top vision-language models reach only 54.2% accuracy on coarse orientation tasks and 33% on granular judgments, with sharp drops on reference-frame shifts and compound rotations.
-
A Comprehensive Evaluation of Deep Learning Object Detection Models on Heterogeneous Edge Devices
SSD MobileNet V1 shows lowest latency and energy but lowest accuracy while YOLOv8 Medium shows highest accuracy at higher cost; Jetson Orin Nano gives best overall balance and accuracy gaps widen with increasing scene complexity.