Detecting Frozen Video in Robot Datasets: Why First/Last Frame Comparison Isn't Enough

说明与用途

机器人数据集中经常出现冻结视频片段，简单的首尾帧比较无法检测到。使用RDA的动作交叉验证检测能够准确识别真实冻结并将其与空闲或降质帧区分开来。

问题

Robotics learning datasets often contain frozen video segments that are not detected by simple first/last frame comparison, leading to false positives and missed failures.

结果

RDA flagged 6 frozen episodes in LIBERO (1.6%) and found no freezes in G1_WBT while correctly identifying degraded quality and idle periods, improving audit accuracy over naive methods.

工作流

• Decode video to low‑res grayscale (64×64)
• Compute adaptive pixel‑difference threshold using noise floor
• Detect continuous frozen segments (≥0.5s)
• Cross‑validate with action data to confirm motion should have occurred

证据

RDA’s video freeze detection works in four stages: ...

来源

0 frozen segments detected across all camera streams

来源