---
title: 'WeatherKITTI: KITTI Weather Benchmarks'
url: https://www.emergentmind.com/topics/weatherkitti
type: topic
---

# WeatherKITTI: KITTI Weather Benchmarks

Searching arXiv for recent papers on WeatherKITTI and closely related KITTI-derived adverse-weather benchmarks.
arxiv_search(query="WeatherKITTI KITTI adverse weather LiDAR benchmark", max_results=10)
WeatherKITTI denotes a set of KITTI-derived adverse-weather benchmarks used to study how autonomous-driving perception pipelines behave when originally clear-weather sensor data are corrupted by fog, snow, rain, or related environmental effects. In the recent literature, the name is not attached to a single canonical dataset; rather, it appears in distinct but overlapping benchmark constructions, most prominently a LiDAR place-recognition benchmark built on KITTI odometry and a large-scale LiDAR denoising dataset derived from SemanticKITTI. Related work extends the same KITTI-based adverse-weather paradigm to rainy lane-depth estimation, 3D single-object tracking, depth completion and denoising, and flooded-road perception [2503.12350] [2408.13802].

## 1. Terminology and scope

A recurrent source of confusion is terminological. One line of work introduces **WeatherKITTI** as one half of the ResLPR benchmark for LiDAR-based place recognition, whereas another introduces **Weather-KITTI** as a large-scale synthetic LiDAR denoising dataset with point-wise weather labels. Both inherit KITTI sensor data and impose controlled adverse-weather corruptions, but they differ in representation, annotation granularity, intended downstream task, and evaluation protocol [2503.12350] [2408.13802].

| Designation | Primary task | Core contents |
|---|---|---|
| WeatherKITTI | LiDAR place recognition | 156,951 range images, clean database C00, corrupted query variants S00/F00/R00, loop-closure labels |
| Weather-KITTI | Point-cloud denoising | 22 SemanticKITTI sequences, raw LiDAR matrices $P_i\in\mathbb{R}^{n_i\times 5}$, automatic point-wise weather labels, 130,656 frames |
| Related KITTI-derived weather resources | Depth, tracking, multi-modal robustness | RainKITTI, KITTI-A, AURORA-KITTI, and broader benchmark proposals including flood data |

This multiplicity matters because results reported under the WeatherKITTI name are not directly interchangeable. A place-recognition score such as Recall@1 or $\mathrm{mSR}_l$ and a denoising score such as mIoU measure different failure modes, even when the underlying weather simulators are similar. This suggests that “WeatherKITTI” is best understood as a benchmark family or naming convention rather than a single immutable corpus.

## 2. Construction principles and weather simulation

In the ResLPR formulation, WeatherKITTI is built from the KITTI odometry dataset’s **64-beam Velodyne HDL-64E** scans. Each raw scan is projected into a **$64\times1920$** 2D range image with two channels per pixel, encoding the normalized range $d$ and intensity $i$. Training sequences for restoration and robustness tuning are KITTI sequences **03–10**, while sequence **00** is used at test time as the clean database and its corrupted variants serve as query sets [2503.12350].

The weather corruption process is explicitly physics-inspired. Fog is modeled through backscatter and attenuation using a hard-target versus soft-target formulation,
\[
\hat{\mathbf p}_f =
\begin{cases}
\bigl(s\,(x,y,z),\,i_{\rm soft}\bigr), & i_{\rm soft}>i_{\rm hard},\\[3pt]
\bigl(x,y,z,\,i_{\rm hard}\bigr), & \text{otherwise},
\end{cases}
\]
with
\[
i_{\rm hard}=i\exp(-2\alpha\,R_0), \qquad
i_{\rm soft}=i\,R_0^2\,\beta\,i_t.
\]
Snowfall follows a scattering-and-occlusion model,
\[
i_{\rm snow}
= T_R + i_{\max}\,f_s\,
\bigl|\,f_o - (1 - R^*/R_{\max})\bigr|^2, \qquad
\hat{\mathbf p}_s = \Bigl(\tfrac{r_s}{\gamma}\,(x,y,z),\,i_{\rm snow}\Bigr),
\]
and rain is generated by Monte Carlo particle simulation, with rain-rate $r_r$ controlling probabilistic occlusion and scattering [2503.12350].

The TripleMixer Weather-KITTI dataset adopts a related but more point-centric construction. It reuses **all 22 SemanticKITTI odometry sequences (00–21)**, collected with a **Velodyne HDL-64E** operating at **64 channels**, **10 Hz**, **$\sim120$ m max range**, **$\sim2$ cm range precision**, and **vertical resolution $\sim0.4^\circ$**. For each sweep, it preserves a matrix
\[
P_i\in\mathbb{R}^{n_i\times 5},
\]
containing $(x,y,z,i,r)$, while also retaining the original timestamp, reflectivity, and full GPS/IMU pose logs. Synthetic weather injection is applied to every frame for **light, moderate, and heavy** snow, fog, and rain, using explicit parameter ranges: snowfall $r_s\in[0.5,1.0]$, $[1.5,2.0]$, $[2.5,3.0]$ mm/h; fog $\beta\in[0.01,0.05]$, $[0.08,0.14]$, $[0.18,0.25]$; rain $r_r\in[1.0,1.5]$, $[1.8,2.4]$, $[2.6,3.0]$ mm/h [2408.13802].

A notable distinction is representational fidelity. ResLPR WeatherKITTI converts the LiDAR stream into range images for restoration and descriptor extraction, whereas TripleMixer Weather-KITTI retains raw point-cloud acquisition information, including range and reflectivity, and automatically generates point-wise weather labels. The former is optimized for loop-closure evaluation; the latter is optimized for denoising and semantic filtering.

## 3. Data organization, labels, and evaluation protocols

For LiDAR place recognition, WeatherKITTI contains **156,951 LiDAR scans** represented as **$64\times1920$** range images across all weather types and severity levels. The test protocol uses sequence **00** clean (**C00**) as the reference database, while the nine corrupted variants—**S00, F00, R00** at severities **1, 2, 3**—serve as query scans. Each corrupted scan inherits the original KITTI pose, and loop-closure ground truth is defined by proximity **within 5 m and $10^\circ$**. The output label for a query scan is the index set of database scans satisfying that loop-closure threshold [2503.12350].

The principal metrics in this setting are **Recall@K**—specifically **R@1** and **R@5**—together with **precision-recall AUC** and **F1 score**, aggregated over R@1 results. ResLPR further introduces the **mean stability rate** $\mathrm{mSR}_l$ as a relative robustness measure:
\[
\text{mSR}_l
= \frac1{3}\sum_{i\in\{\text{snow,fog,rain}\}}
\frac{\sum_{s=1}^3\bigl[\text{AUC}_{i,s}+\text{F1}_{i,s}+\text{R@1}_{i,s}+\text{R@5}_{i,s}\bigr]}
{3\,\bigl[\text{AUC}_{\rm clean}+\text{F1}_{\rm clean}+\text{R@1}_{\rm clean}+\text{R@5}_{\rm clean}\bigr]}\,.
\]
This metric is designed to express degradation relative to clean-weather performance, rather than reporting only absolute retrieval scores [2503.12350].

In the denoising-oriented Weather-KITTI dataset, scale and annotation differ substantially. The corpus contains **130,656 frames**, with approximately **14,000 frames per (type, severity)** slice, and an average of **$\approx100,000$–$120,000$ points per frame**. The dataset reports global totals of **139 M snow points**, **383 M fog points**, and **89 M rain points**. Weather noise is labeled at point level using category IDs **110** for snow noise, **111** for fog noise, and **112** for rain noise, while original SemanticKITTI classes remain in **0–29**. The per-frame file set includes `.bin` point clouds, `.label` files, `pose.csv`, and `metadata.json` with LiDAR sensor parameters such as focal offset, focal slope, and pulse width $\tau_H$ [2408.13802].

The denoising evaluation protocol uses train/validation/test splits partitioned per weather modality. Its primary metrics are
\[
\mathrm{Precision}=\frac{TP}{TP+FP},\qquad
\mathrm{Recall}=\frac{TP}{TP+FN},
\]
\[
F_1=\frac{2\cdot(\mathrm{Precision}\cdot\mathrm{Recall})}{\mathrm{Precision}+\mathrm{Recall}},\qquad
\mathrm{mIoU}=\frac{TP}{TP+FP+FN}.
\]
The paper notes that Chamfer Distance or per-point SNR could also be reported in principle, but the main protocol is classification-based [2408.13802].

## 4. Restoration and denoising architectures

Within the ResLPR benchmark, adverse weather is not only a source of evaluation corruption but also the target of an explicit preprocessor. **ResLPRNet** is a plug-and-play LiDAR range-image restoration network intended to remove weather-induced noise and fill missing returns before descriptor extraction. Its backbone contains **WaveTransformer (WAT) Blocks** in both encoder and decoder. These blocks first apply a discrete wavelet transform,
\[
\mathbf F\to \{\mathbf F_{LL},\mathbf F_{LH},\mathbf F_{HL},\mathbf F_{HH}\}=\mathrm{DWT}(\mathbf F),
\]
then reconstruct wavelet features by
\[
\mathbf F_w=\mathrm{ConvTrans}(\mathrm{Concat}(\mathbf F_{LL},\dots,\mathbf F_{HH})).
\]
Feature mixing combines spatial self-attention and grouped-convolution channel mixing, after which a transformer layer fuses $\mathbf F_w$ with the mixed features to capture global context. Decoder stages further include **ContextGuide (CTG) Blocks** with learnable context embeddings $\mathbf{CE}$ and softmax-weighted aggregation,
\[
\mathbf W = \mathrm{Softmax}(\mathrm{FC}(\mathrm{GAP}(\mathbf F))), \qquad
\mathbf F_{cb}=\mathrm{MLP}(\sum_k W_k\,CE_k),
\]
to adapt to weather-type-specific degradation. Skip connections preserve spatial detail, and the restoration objective is
\[
L=\frac1N\sum_{j=1}^N\bigl(|d_j-\hat d_j|+|i_j-\hat i_j|\bigr).
\]
Training uses Adam with **lr = $1\times10^{-4}$** for **120 epochs** on **$32\times480$** random crops with horizontal and vertical flips [2503.12350].

The TripleMixer line of work addresses the same weather corruption problem at point-cloud level. **TripleMixer** is composed of a **Geometry Mixer Layer**, a **Frequency Mixer Layer**, and a **Channel Mixer Layer**, respectively designed to capture geometric spatial information, extract multi-scale frequency information, and enhance multi-channel feature information of point clouds. The evaluation includes statistical filters—**SOR, ROR, DSOR, DROR**—general semantic segmenters adapted to binary classification—**SalsaNext, Cylinder3D**—and specialized denoisers such as **WeatherNet, 4DenoiseNet, 3D-OutDet** [2408.13802].

These two architectures occupy complementary positions in the pipeline. ResLPRNet is range-image restoration tightly coupled to place recognition, whereas TripleMixer is point-cloud denoising with semantic weather labels. The shared premise is that adverse-weather robustness can be improved by front-end correction rather than by modifying only the downstream descriptor or segmentation network.

## 5. Empirical performance and task-specific extensions

On the ResLPR WeatherKITTI benchmark, clean-weather place-recognition baselines—**Scan Context (SC)**, **CVTNet**, and **LPSNet**, all pretrained on clean KITTI—typically achieve **R@1 $>0.8$** and **$F1\approx0.8$–$0.9$** on clean KITTI 00. Under moderate weather corruption without preprocessing, however, **R@1** often falls **below 0.2** and **F1** to **$<0.1$**. For corrupted scans only, SC records approximately **AUC $\approx0.14$**, **F1 $\approx0.18$**, and **R@1 $\approx0.16$** on snow, with similarly low values on fog and rain; overall $\mathrm{mSR}_l$ is approximately **0.31** for SC, **0.19** for CVTNet, and **0.36** for LPSNet. With ResLPRNet preprocessing, the reported gains are much larger: **SC+ResLPR** reaches snow **AUC $\to0.80$ (+0.66)**, **F1 $\to0.77$**, **R@1 $\to0.80$**, **R@5 $\to0.84$**, with **$\mathrm{mSR}_l=0.95$**; **CVTNet+ResLPR** reaches snow **AUC $\to0.61$ (+0.59)**, **R@1 $\to0.79$**, with **$\mathrm{mSR}_l=0.79$**; and **LPSNet+ResLPR** reaches snow **AUC $\to0.52$** versus **0.03** corrupted and **R@1 $\to0.83$**, with **$\mathrm{mSR}_l=0.92$**. Comparative preprocessing methods, including **WeatherNet** and **TripleMixer**, provide only modest gains, with $\mathrm{mSR}_l$ up to **$\sim0.55$**, below ResLPRNet’s **$\sim0.9+$** stability [2503.12350].

On the denoising side, TripleMixer reports that Weather-KITTI supports state-of-the-art denoising evaluation across snow, fog, and rain, with **average mIoU = 96.31%** and an improvement over prior methods by **at least 2–16% in absolute terms**. Because the dataset preserves all original clean scene points and labels only introduced weather returns, the task is explicitly formulated as identifying and suppressing synthetic weather noise while maintaining scene structure [2408.13802].

The same adverse-weather KITTI logic has been extended to other tasks. **RainKITTI** synthesizes rainy lane images from KITTI left-camera frames using the **RCFLane** pipeline, which sequentially adds a rain layer, global darkening, and depth-dependent fog:
\[
O_1(x)=\alpha(x)\,O(x)+\beta\,R(x),\qquad
O_2(x)=\gamma\,O_1(x)+(1-\gamma)\,D(x),
\]
\[
\hat O(x)=O_2(x)\,\tau(x)+A(1-\tau(x)).
\]
The dataset contains **820** original KITTI frames and **820** synthesized rainy frames at **$1238\times374$**, split into **715** training and **105** test images. The associated **DLKPN** restores rainy inputs using two predicted pixel-wise kernels, and its reconstruction score reaches **SSIM 0.941** and **PSNR 30.601** with **152.9 ms** runtime; in depth estimation, **KPNDepth(D)** reports **Abs Rel 0.066**, **Sq Rel 0.447**, **RMSE 2.897**, **RMSE log 0.136**, and **$\delta<1.25 = 0.941$** [2405.09964].

For LiDAR tracking, **KITTI-A** applies synthetic **fog, rain, and snow** to the KITTI tracking split at **five discrete intensity levels per weather**, producing **15 variants** of each original sequence. It retains **Car** and **Pedestrian** categories and reports **1,800 sequences / 96,360 frames** for Car and **930 sequences / 91,320 frames** for Pedestrian. Evaluation uses **Success**, **Precision**, and a **Degradation Rate**
\[
\mathrm{DR}^{S}_{i,j}
=
\frac{S_c-S_{i,j}}{S_c},
\]
where $S\in\{\text{success, precision}\}$ [2501.07133].

A broader multimodal extension appears in **AURORA-KITTI**, which introduces **82,177 aligned RGB–LiDAR pairs** with metric dense-depth ground truth across **{day, night} $\times$ {clear, fog, rain, snow}**, with three severity levels for fog, rain, and snow. It defines **Depth Completion and Denoising (DCD)** as a unified task and reports that **DDCD** achieves **RMSE 1,799.55 mm**, **MAE 566.31 mm**, **iRMSE 4.13**, and **iMAE 1.93** on AURORA-KITTI zero-shot evaluation [2603.14701].

The most explicit real-world counterpart to these synthetic weather resources is **FRED**, a multimodal flooded-road dataset released in both KITTI-style and RTMaps formats. It comprises **10 sequences** from **five** locations, with approximately **5,340** time-synchronized samples, **FLIR** images, **Ouster OS1-64** LiDAR, and **IMU/GNSS** data. Its authors position it as the **“flood” component of a broader WeatherKITTI benchmark**, thereby extending the weather-benchmark idea beyond simulated rain, fog, and snow [2605.22018].

## 6. Significance, misconceptions, and limitations

The main scientific role of WeatherKITTI-style resources is controlled robustness evaluation. By parameterizing severity and preserving either loop-closure labels, point-wise weather classes, or dense depth targets, these datasets make it possible to isolate failure modes induced by attenuation, backscatter, scattering, occlusion, and spurious returns. This control is especially clear in the ResLPR benchmark’s use of clean reference databases plus corrupted query sets, and in TripleMixer’s preservation of all original clean scene points alongside exact synthetic-noise labels.

A common misconception is that WeatherKITTI names a single benchmark with a single protocol. The literature instead presents multiple task-specific instantiations: range-image place recognition, point-cloud denoising, lane depth estimation, 3D tracking, and unified depth completion. As a result, benchmark numbers are not directly comparable across papers because the modalities, annotation schemas, and metrics differ. This suggests that precise citation of the variant—ResLPR WeatherKITTI, TripleMixer Weather-KITTI, or a related extension—is methodologically important.

Another limitation concerns the relation between synthetic and real adverse weather. The dominant WeatherKITTI constructions are simulation-based, using physics-inspired or physics-based weather injection on clear KITTI data. Parallel work on real-weather transfer and real-world collection indicates why this distinction matters. In source-only cross-weather LiDAR segmentation, training on SemanticKITTI and evaluating on the real-weather **SemanticSTF** benchmark yields **24.4 mIoU** for a source-only baseline, **36.3 mIoU** for data-centric augmentation, **38.9 mIoU** for class-centric regularization, and **39.15 mIoU** for a geometry-aware regularization method, showing that robustness still depends on transfer beyond synthetic corruption alone [2511.01250]. Likewise, AURORA-KITTI explicitly concludes that **weather-aware, physically consistent data contributes more to robustness than architectural modifications alone**, and FRED supplies real flooded-road data rather than simulated water hazards [2603.14701] [2605.22018].

Taken together, these developments position WeatherKITTI not merely as a dataset name but as a methodological template: start from a standard KITTI-derived corpus, inject or collect adverse-weather conditions with controlled semantics, and evaluate robustness in a task-specific manner. The continuing proliferation of related benchmarks suggests an ongoing shift from clear-weather accuracy toward explicit all-weather stress testing across localization, recognition, denoising, tracking, segmentation, and depth reconstruction.

Source: https://www.emergentmind.com/topics/weatherkitti