Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hosseinzadeh, Mehdi, Wong, King Hang, Dayoub, Feras
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2604.07034
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914457955860480
author Hosseinzadeh, Mehdi
Wong, King Hang
Dayoub, Feras
author_facet Hosseinzadeh, Mehdi
Wong, King Hang
Dayoub, Feras
contents We present KITE, a training-free, keyframe-anchored, layout-grounded front-end that converts long robot-execution videos into compact, interpretable tokenized evidence for vision-language models (VLMs). KITE distills each trajectory into a small set of motion-salient keyframes with open-vocabulary detections and pairs each keyframe with a schematic bird's-eye-view (BEV) representation that encodes relative object layout, axes, timestamps, and detection confidence. These visual cues are serialized with robot-profile and scene-context tokens into a unified prompt, allowing the same front-end to support failure detection, identification, localization, explanation, and correction with an off-the-shelf VLM. On the RoboFAC benchmark, KITE with Qwen2.5-VL substantially improves over vanilla Qwen2.5-VL in the training-free setting, with especially large gains on simulation failure detection, identification, and localization, while remaining competitive with a RoboFAC-tuned baseline. A small QLoRA fine-tune further improves explanation and correction quality. We also report qualitative results on real dual-arm robots, demonstrating the practical applicability of KITE as a structured and interpretable front-end for robot failure analysis. Code and models are released on our project page: https://m80hz.github.io/kite/
format Preprint
id arxiv_https___arxiv_org_abs_2604_07034
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle KITE: Keyframe-Indexed Tokenized Evidence for VLM-Based Robot Failure Analysis
Hosseinzadeh, Mehdi
Wong, King Hang
Dayoub, Feras
Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
We present KITE, a training-free, keyframe-anchored, layout-grounded front-end that converts long robot-execution videos into compact, interpretable tokenized evidence for vision-language models (VLMs). KITE distills each trajectory into a small set of motion-salient keyframes with open-vocabulary detections and pairs each keyframe with a schematic bird's-eye-view (BEV) representation that encodes relative object layout, axes, timestamps, and detection confidence. These visual cues are serialized with robot-profile and scene-context tokens into a unified prompt, allowing the same front-end to support failure detection, identification, localization, explanation, and correction with an off-the-shelf VLM. On the RoboFAC benchmark, KITE with Qwen2.5-VL substantially improves over vanilla Qwen2.5-VL in the training-free setting, with especially large gains on simulation failure detection, identification, and localization, while remaining competitive with a RoboFAC-tuned baseline. A small QLoRA fine-tune further improves explanation and correction quality. We also report qualitative results on real dual-arm robots, demonstrating the practical applicability of KITE as a structured and interpretable front-end for robot failure analysis. Code and models are released on our project page: https://m80hz.github.io/kite/
title KITE: Keyframe-Indexed Tokenized Evidence for VLM-Based Robot Failure Analysis
topic Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.07034