Pith Number

pith:VW7OKKUD

pith:2026:VW7OKKUDXU4I2C2O64IQIT6TPO

not attested not anchored not stored refs resolved

Weakly-Supervised Spatiotemporal Anomaly Detection

Mubarak Shah, Praveen Tirupattur, Urvi Gianchandani

A weakly supervised classifier with multiple instance ranking loss can localize video anomalies in both space and time from video-level labels alone.

arxiv:2605.13746 v1 · 2026-05-13 · cs.CV · cs.AI

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{VW7OKKUDXU4I2C2O64IQIT6TPO}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

Features extracted from normal or anomalous video clips are used to determine anomaly scores for spatiotemporal regions based on a classifier and multiple instance ranking loss, enabling detection on the UCF Crime2Local Dataset.

C2weakest assumption

That video-level labels alone, combined with a standard MIL ranking loss, are sufficient to localize anomalies both spatially within frames and temporally within clips without additional supervision or post-hoc selection.

C3one line summary

A multiple instance learning approach with ranking loss localizes spatiotemporal anomalies in videos using only video-level normal/anomalous labels on the UCF Crime2Local dataset.

References

4 extracted · 4 resolved · 2 Pith anchors

[1] It is made up of both normal and anomalous surveillance videos 1900

[2] Anomaly Locality in Video Surveillance 1901 · arXiv:1901.10364

[3] A revisit of sparse coding based anomaly detection in stacked rnn framework 2017

[4] Future Frame Prediction for Anomaly Detection -- A New Baseline · arXiv:1712.09867

Receipt and verification

First computed	2026-05-18T02:44:16.411577Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

adbee52a83bd388d0b4ef711044fd37b8f17631cfc401923fcd7d292f2b87ad5

Aliases

arxiv: 2605.13746 · arxiv_version: 2605.13746v1 · doi: 10.48550/arxiv.2605.13746 · pith_short_12: VW7OKKUDXU4I · pith_short_16: VW7OKKUDXU4I2C2O · pith_short_8: VW7OKKUD

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/VW7OKKUDXU4I2C2O64IQIT6TPO \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: adbee52a83bd388d0b4ef711044fd37b8f17631cfc401923fcd7d292f2b87ad5

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "afbf576d4124999c597b4db60dd868d99d428ba28e66a06f77dbd05e9f607f96",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-13T16:28:01Z",
    "title_canon_sha256": "f2c7ae9f4a961b06702d2d618e6f17bbac3408ee5b52b8231d6ef12a598d96fb"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13746",
    "kind": "arxiv",
    "version": 1
  }
}