Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 62 of the 62 references it cites were found in a registry or an index; 4 of the 17 sources its claims cite were checked for support.

Unsupervised Active Visual Search With Monte Carlo Planning Under Uncertain Detections

Open access

Abstract

We propose a solution for Active Visual Search of objects in an environment, whose 2D floor map is the only known information. Our solution has three key features that make it more plausible and robust to detector failures compared to state-of-the-art methods: i) it is unsupervised as it does not need any training sessions. ii) During the exploration, a probability distribution on the 2D floor map is updated according to an intuitive mechanism, while an improved belief update increases the effectiveness of the agent's exploration. iii) We incorporate the awareness that an object detector may fail into the aforementioned probability modelling by exploiting the success statistics of a specific detector. Our solution is dubbed POMP-BE-PD (Pomcp-based Online Motion Planning with Belief by Exploration and Probabilistic Detection). It uses the current pose of an agent and an RGB-D observation to learn an optimal search policy, exploiting a POMDP solved by a Monte-Carlo planning approach. On the Active Vision Dataset Benchmark, we increase the average success rate over all the environments by a significant 35 % while decreasing the average path length by 4 % with respect to competing methods. Thus, our results are state-of-the-art, even without any training procedure.

Overview

CoChat Summary

TL;DR

POMP-BE-PD is an unsupervised active visual search method that plans online with POMCP in a known 2D floor map, using probabilistic object detection and an improved belief update. On the Active Vision Database benchmark it reports higher success rates than competing methods and its prior POMP formulation without any training procedure.

Key findings

  • With the real detector, POMP-BE-PD reaches average SR 0.54, APL 39.44, and SPL 0.47 across the tested AVDB scenes, versus POMP’s 0.40, 41.13, and 0.50; the paper reports this as a 35% relative increase in success rate and a 4% decrease in average path length, or a 14% absolute SR increase in the conclusion.
  • Probabilistic detection reduces false positives by about 32%, and the docking strategy reduces docking error by about 35.7% (Fig. 8).
  • With ground-truth annotations, the improved belief update alone (POMP-BE) raises medium-scene SR from 0.76 to 0.85 and lowers APL from 29.64 to 25.50; hard-scene SR rises from 0.48 to 0.70.
  • On the GT-object protocol in Table 1, POMP-BE-PD achieves average SR 0.90 versus POMP’s 0.76, with APL 18.10 versus 17.07 and SPL 0.72 versus 0.75.

Methods

  • The task is modeled as a POMDP solved by Partially Observable Monte Carlo Planning (POMCP); the agent uses its pose and RGB-D observations, with a pose graph, candidate object locations, and a visibility matrix derived from the known 2D floor map.
  • Detection uses the Target Driven Instance Detector (TDID), keeping detections with score > 0.9; class-specific F1 statistics feed a probabilistic detection vector and Bayesian update over object locations, with exploration terminating when a visible location’s probability exceeds threshold τ.
  • Experiments use the Active Vision Database benchmark with 8 scenes (2 easy, 3 medium, 3 hard), metrics SR, APL, and SPL, a 200-step limit, and comparisons against Random Walk, EAT, DQN, DQN-TAM, and POMP.

Limitations

  • Stated: no official code for published comparison methods was available, so comparisons follow the protocol from [9], which uses only GT annotations and a limited number of scenes; DQN and DQN-TAM use only two scenes and a different object subset, making their average column not meaningful for fair comparison.
  • Evident from the text: the method assumes a known 2D floor map and relies on an off-the-shelf object detector whose class-specific F1 statistics must be available; performance depends on threshold τ and simulation budget.
  • Evident from the text: evaluation is limited to the AVDB benchmark and selected scenes, and probabilistic detection can increase APL and reduce SPL in some settings.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1109/tpami.2024.3451994

Cites

62 of 62 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 Valid · 4 Not valid · 7 Not checked · 31 Own results and interpretations

14 more claims

Create a free account to see all 42 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 19
  • Probably not a substantial claim. 5
  • The quote only partly states this claim. 2

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 60 references

Research works printed but not linked: 1

  1. Mazzi, Active generation of logical rules for POMCP shielding, 1696

Cited by

See all 8

Metadata

Authors
Taioli, F., Giuliari, F., Wang, Y., Berra, R., Castellini, A., Bue, A. D., Farinelli, A., Cristani, M., & Setti, F.
Affiliations
  • University of Verona (IT)
  • Italian Institute of Technology (IT)
  • Fondazione Bruno Kessler (IT)
Venue
IEEE Transactions on Pattern Analysis and Machine Intelligence
Published
2024-12
Volume
46
Issue
12
Pages
11047-11058
ISSN
  • 0162-8828 (ISSN-L)
  • 2160-9292
  • 1939-3539
OpenAlex
W4401990919
Cited by
8
References
60
Citations by year
2026: 5 · 2025: 2
FWCI
1.03
Type
journal article
Language
en
Open access
Locations
  • IEEE Transactions on Pattern Analysis and Machine Intelligence (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • PORTO@Iris (submittedVersion, cc-by)
  • arXiv (Cornell University)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Robotics and Sensor-Based Localization
  • Physical Sciences › Engineering › Control and Systems Engineering › Robot Manipulation and Learning
Keywords
object detection, unsupervised learning, partially observable Markov decision processes, belief updating, sensor faults, probabilistic detection, online motion planning, floor maps, RGB-D sensing

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-07, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-07 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Unsupervised Active Visual Search With Monte Carlo Planning Under Uncertain Detections. https://app.cochat.ai/papers/10.1109/tpami.2024.3451994
The paper
Taioli, F., Giuliari, F., Wang, Y., Berra, R., Castellini, A., Bue, A. D., Farinelli, A., Cristani, M., & Setti, F. (2024). Unsupervised Active Visual Search With Monte Carlo Planning Under Uncertain Detections. IEEE Transactions on Pattern Analysis and Machine Intelligence. https://doi.org/10.1109/tpami.2024.3451994

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.