---
title: 'SO-HOTA: Small-Object-Aware HOTA Metric'
url: https://www.emergentmind.com/topics/so-hota
type: topic
---

# SO-HOTA: Small-Object-Aware HOTA Metric

SO-HOTA is a small-object-aware variant of HOTA for multi-object tracking, introduced in the SMOT4SB challenge to evaluate small multi-object tracking in UAV video where birds often occupy only a few dozen pixels and IoU-based matching becomes unstable [2507.12832]. It preserves HOTA’s balanced treatment of detection and association, but replaces IoU-driven matching with Dot Distance (DotD), a center-distance-based similarity designed to remain informative under tiny spatial displacements that would otherwise cause overlap-based scores to collapse [2507.12832]. In the cited literature, SO-HOTA is therefore a metric, not a tracker: it is intended to rank SMOT systems more faithfully in regimes where standard HOTA’s localization sensitivity becomes a dominant source of evaluation error [2507.12832].

## 1. Conceptual origin and problem setting

SO-HOTA arises from a specific failure mode of conventional MOT evaluation on very small targets. In SMOT4SB, birds are frequently smaller than \(32\times 32\) pixels, and many are around \(16\times 16\) or similarly small scales; at that size, a localization error of only a few pixels can radically change IoU even when the prediction remains visually close to the target [2507.12832]. The motivating example is explicit: for a \(16\times 16\) box, an 8-pixel horizontal or vertical shift already reaches the conventional IoU threshold boundary of \(0.5\), and once the shift exceeds half the box size, IoU becomes zero [2507.12832].

The paper identifies two pathologies of overlap-based evaluation in this regime. First, the metric becomes **overly sensitive**: trivial-looking localization errors can produce a large score drop. Second, IoU becomes **non-discriminative once overlap is lost**: a prediction that is barely non-overlapping and one that is far away both receive IoU \(=0\) [2507.12832]. In UAV bird tracking, where both camera motion and bird motion are unconstrained in 3D, near-but-non-overlapping predictions occur frequently, so a pure IoU criterion can undervalue tracking quality [2507.12832].

This setting explains why SO-HOTA was proposed inside a challenge paper rather than a general MOT benchmark paper. The SMOT4SB challenge introduces three main contributions: the SMOT4SB dataset with 211 UAV video sequences and 108,192 annotated frames, SO-HOTA as the primary ranking metric, and a competitive challenge with 78 participants and 308 submissions [2507.12832]. The metric is thus inseparable from the small-object tracking regime for which it was designed.

## 2. Relation to standard HOTA

SO-HOTA is structurally derived from HOTA, the Higher Order Tracking Accuracy metric introduced to balance detection and association while incorporating localization through threshold averaging [2009.07736]. At a localization threshold \(\alpha\), standard HOTA decomposes as
$$
\mathrm{HOTA}_{\alpha}=\sqrt{\mathrm{DetA}_{\alpha}\cdot \mathrm{AssA}_{\alpha}},
$$
with
$$
\mathrm{DetA}_{\alpha}=\frac{|\mathrm{TP}|}{|\mathrm{TP}|+|\mathrm{FN}|+|\mathrm{FP}|},
$$
and
$$
\mathrm{AssA}_{\alpha}=\frac{1}{|\mathrm{TP}|}\sum_{c\in\{\mathrm{TP}\}} \mathcal{A}(c),
$$
where \(\mathcal{A}(c)\) is a Jaccard-style association score defined from \(\mathrm{TPA}(c)\), \(\mathrm{FNA}(c)\), and \(\mathrm{FPA}(c)\) over the matched predicted and ground-truth trajectories [2009.07736]. Final HOTA is then averaged over thresholds \(\alpha \in \{0.05,0.1,\ldots,0.95\}\) [2009.07736].

The essential design choice in SO-HOTA is not to discard this decomposition. Instead, the paper keeps HOTA’s “optimization structure for detection, association, and localization accuracy,” but changes the similarity notion used at the matching stage from IoU to DotD [2507.12832]. This means SO-HOTA is best understood as a HOTA-family metric with a different admissibility criterion for true positives, rather than as a wholly unrelated evaluation framework.

That continuity matters methodologically. HOTA was introduced precisely because prior metrics such as MOTA and IDF1 overemphasized different parts of the tracking problem, whereas HOTA explicitly balances detection and association [2009.07736]. SO-HOTA preserves that balance, but adapts the underlying geometric primitive to the small-object regime [2507.12832].

## 3. Formal definition and matching procedure

The core similarity in SO-HOTA is Dot Distance:
$$
\mathrm{DotD}(A,B)=e^{-D/S},
$$
where
$$
D=\sqrt{(x_A-x_B)^2+(y_A-y_B)^2},
$$
and \(S\) is the average size of all objects in the dataset, described as the square root of average box area over all labeled objects [2507.12832]. DotD therefore measures center proximity relative to a dataset-level characteristic object scale.

A predicted box and a ground-truth box are considered a true positive at threshold \(\alpha\) when
$$
\mathrm{DotD}(A,B)\ge \alpha.
$$
The paper states that SO-HOTA “follows the same matching procedure as HOTA, using the Hungarian algorithm to assign predicted and ground-truth detections in a one-to-one manner,” with DotD used as the similarity measure instead of IoU [2507.12832]. The practical consequence is that the matched set, and therefore the downstream TP, FP, FN, TPA, FNA, and FPA counts, are all DotD-defined rather than IoU-defined.

For detection accuracy, the paper defines SO-DetA by replacing IoU-based true positives with DotD-based true positives [2507.12832]. For association accuracy, it likewise defines SO-AssA by computing the HOTA-style association score over the DotD-defined matched set [2507.12832]. The per-threshold metric is then
$$
\mathrm{SO\text{-}HOTA}_{\alpha}=\sqrt{\mathrm{SO\text{-}DetA}_{\alpha}\times \mathrm{SO\text{-}AssA}_{\alpha}},
$$
and the final score is
$$
\mathrm{SO\text{-}HOTA}=\frac{1}{19}\sum_{\alpha\in\{0.05,0.1,\ldots,0.95\}} \mathrm{SO\text{-}HOTA}_{\alpha}.
$$
This threshold averaging mirrors standard HOTA’s multi-threshold philosophy, but substitutes DotD for IoU in the matching backbone [2507.12832].

The metric therefore changes HOTA primarily at the **matching/similarity stage**. A plausible implication is that SO-HOTA should be viewed as a DotD-based adaptation of HOTA’s DetA/AssA machinery, not as a new association formalism. The paper is explicit that it does not provide an explicit separate formula for a modified localization-accuracy term, even though it says the method preserves HOTA’s overall optimization structure [2507.12832].

## 4. Behavior in the SMOT4SB challenge

SO-HOTA was the primary ranking criterion in the SMOT4SB challenge: teams were ranked by their highest achieved SO-HOTA score, with public evaluation through CodaBench and private evaluation performed by the organizers on submitted code [2507.12832]. In addition to SO-HOTA, the challenge also reported conventional HOTA, MOTA, IDF1, Mostly Tracked, and Mostly Lost for reference, alongside the SO-HOTA suite metrics SO-HOTA, SO-DetA, SO-AssA, SO-DetRe, and SO-DetPr [2507.12832].

The reported scores illustrate the intended effect of the new metric. On the private test set, the winning DL Team achieved **SO-HOTA = 50.59** and **HOTA = 36.74**, while the baseline scored **SO-HOTA = 9.90** and **HOTA = 6.51** [2507.12832]. On the public test set, DL Team achieved **SO-HOTA = 54.90** and **HOTA = 41.67** [2507.12832]. The challenge paper interprets the persistent gap between SO-HOTA and standard HOTA as evidence that conventional overlap-based HOTA is harsher in the tiny-object regime, whereas SO-HOTA retains discriminative power under slight localization offsets [2507.12832].

The paper’s synthetic displacement analysis clarifies this behavior. When a prediction is vertically displaced by \(x\) pixels from the ground truth for a box of size \(d\), IoU falls sharply and reaches zero when overlap disappears, whereas DotD decays smoothly with center distance [2507.12832]. The corresponding HOTA-versus-SO-HOTA comparison shows HOTA reacting dramatically even to 1-pixel displacement and collapsing once overlap is lost, while SO-HOTA continues to provide graded evaluation based on proximity [2507.12832]. This is the metric’s main empirical justification.

The challenge results also align with tracker design trends. The paper notes that several top methods moved away from pure IoU for association, instead using adaptive similarity, distance penalties, DIoU, box expansion, or motion compensation [2507.12832]. This suggests that evaluation and algorithm design converged on the same operational insight: in tiny-object tracking, spatial tolerance is often necessary because overlap may vanish even for semantically correct correspondences.

## 5. Scope, neighboring literature, and recurrent misreadings

The most common misreading is to treat SO-HOTA as a generic synonym for “HOTA-style tracking” or for strong association performance. That is incorrect. Several MOT papers are relevant to SO-HOTA-oriented literature searches because they report strong HOTA or AssA results, but they do **not** define SO-HOTA. SearchTrack, for example, is an online MOTS/MOT method whose relevance comes from reporting HOTA, DetA, and AssA and from demonstrating large association gains, especially through motion-aware features, but it does not propose a new higher-order metric [2210.16572]. Deep HM-SORT is similarly centered on sports MOT and reports strong HOTA and AssA, yet it does not mention SO-HOTA explicitly [2406.12081]. RTAT ranks first on MOT17 and MOT20 in several HOTA-family metrics through a two-stage association design, but again its contribution is a tracker, not an evaluation metric [2408.07344].

A second misreading is to attribute SO-HOTA to sparse-label long-horizon tracking benchmarks. The multi-camera traffic paper “So you think you can track?” uses standard HOTA terminology and modifies the detection term to
$$
\mathrm{DetA}^{*}_{\alpha}=\frac{\mathrm{TP}}{\mathrm{TP}+\mathrm{FN}},
$$
because false positives cannot be counted under sparse GPS-based ground truth, but it explicitly does **not** define SO-HOTA [2309.07268]. That work motivates a recall-oriented HOTA adaptation for sparse labels, not the DotD-based small-object metric introduced in SMOT4SB.

A third source of confusion is acronym overload. “HOTA” does not uniquely refer to the MOT metric family.

| Term | Meaning | Relation to SO-HOTA |
|---|---|---|
| HOTA | Higher Order Tracking Accuracy | Metric foundation [2009.07736] |
| SO-HOTA | DotD-based small-object HOTA variant | Topic of SMOT4SB [2507.12832] |
| \(\mathrm{DetA}^{*}_{\alpha}\) usage | Recall-oriented HOTA modification under sparse labels | Not named SO-HOTA [2309.07268] |
| HOTA-FedGradNorm | Hierarchical Over-the-Air FedGradNorm | Unrelated acronym [2212.07414] |
| HOTA in OT | Hamiltonian framework for Optimal Transport Advection | Unrelated acronym [2507.17513] |

This terminological ambiguity matters in bibliographic practice. A literature search for “SO-HOTA” can return work on HOTA-family tracking metrics, traffic datasets with modified HOTA components, or entirely unrelated “HOTA” acronyms in federated learning and optimal transport. In the tracking literature represented here, only the SMOT4SB challenge paper explicitly introduces SO-HOTA as a named metric [2507.12832].

## 6. Limitations and methodological implications

The challenge paper does not present SO-HOTA as a universal replacement for HOTA. Several limitations are either stated directly or are structurally apparent. First, SO-HOTA still depends on bounding boxes and still inherits HOTA’s framework, so it does not address annotation ambiguity in object extent or uncertainty beyond box-level correspondence [2507.12832]. Second, DotD depends only on center distance. Two boxes with similar centers but different sizes or aspect ratios can therefore score highly, which the paper treats as an acceptable trade-off for tiny birds whose center location is more informative than exact extent [2507.12832].

Third, the normalization scale \(S\) is computed globally from all labeled boxes in the dataset, not per-image, per-sequence, or per-instance [2507.12832]. This makes the metric simple and reproducible, but it also implies that heterogeneous scale distributions may not be perfectly represented by a single characteristic size. Fourth, although the paper states that SO-HOTA preserves HOTA’s optimization structure, it does not formalize a separate SO-localization term; implementers seeking a full analogue of every HOTA submetric would therefore need to infer or design that component themselves [2507.12832].

These limitations imply a narrower but clearer interpretation. SO-HOTA is most appropriately understood as a **small-object-aware HOTA**: it preserves HOTA’s balanced treatment of detection and association, but replaces IoU with DotD in the matching and thresholding logic so that tiny spatial displacements do not cause catastrophic evaluation changes [2507.12832]. Within that scope, it is not merely a benchmark convenience. It expresses a specific evaluative thesis: for SMOT in UAV bird tracking, especially when boxes are very small and motion is highly dynamic, spatial proximity is often a better primitive than overlap for deciding whether a tracker is “close enough” to deserve credit [2507.12832].

Source: https://www.emergentmind.com/topics/so-hota