Pith. sign in

REVIEW 6 cited by

Re-ID done right: towards good practices for person re-identification

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1801.05339 v1 pith:DN5LHLMY submitted 2018-01-16 cs.CV

classification cs.CV
keywords personre-identificationdeeppracticesapproacharchitectureauxiliarycomponents
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Training a deep architecture using a ranking loss has become standard for the person re-identification task. Increasingly, these deep architectures include additional components that leverage part detections, attribute predictions, pose estimators and other auxiliary information, in order to more effectively localize and align discriminative image regions. In this paper we adopt a different approach and carefully design each component of a simple deep architecture and, critically, the strategy for training it effectively for person re-identification. We extensively evaluate each design choice, leading to a list of good practices for person re-identification. By following these practices, our approach outperforms the state of the art, including more complex methods with auxiliary components, by large margins on four benchmark datasets. We also provide a qualitative analysis of our trained representation which indicates that, while compact, it is able to capture information from localized and discriminative regions, in a manner akin to an implicit attention mechanism.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Organizational Control Layer: Governance Infrastructure at the Execution Boundary of LLM Agent Systems

    cs.MA 2026-06 unverdicted novelty 6.0 of 10

    OCL is a governance layer for LLM agents that cuts unsafe executions from 88% to near-zero and raises valid success from 12% to 96% in adversarial buyer-seller negotiations across frontier LLMs.

  2. ROGLE: Robust Global-Local Alignment with Automated Region Supervision for Text-Based Person Search

    cs.CV 2026-06 unverdicted novelty 6.0 of 10

    ROGLE introduces automated pseudo region-sentence pairs via RSM and multi-granular learning to boost fine-grained alignment in text-based person search, plus the P-VLG benchmark with over 100k annotated regions.

  3. Beyond Walking: A Large-Scale Image-Text Benchmark for Text-based Person Anomaly Search

    cs.CV 2024-11 reject novelty 6.0 of 10

    Introduces the Pedestrian Anomaly Behavior (PAB) benchmark and a Cross-Modal Pose-aware model for retrieving pedestrians from text descriptions of normal or anomalous actions, reporting 84.93% R@1.

  4. HorNet: A Hierarchical Offshoot Recurrent Network for Improving Person Re-ID via Image Captioning

    cs.CV 2019-08 conditional novelty 6.0 of 10

    A hierarchical gated recurrent network that fuses image features with generated text captions improves person re-identification on three benchmark datasets, including one with no human captions.

  5. Fine-Grained Action Retrieval Through Multiple Parts-of-Speech Embeddings

    cs.CV 2019-08 conditional novelty 6.0 of 10

    Disentangling captions by parts of speech into separate learned embeddings improves cross-modal text-video retrieval for fine-grained actions.

  6. Person detection and re-identification in open-world settings of retail stores and public spaces

    cs.CV 2025-05 reject novelty 2.0 of 10

    A demo of off-the-shelf person detection and re-identification on an OAK-D camera in retail and public spaces, with only qualitative results.

Pith tools